R wo

ijOi'i UWi 4i<lS iiiflcing Paper

Library ™ " Overseas Development Institute

FOR REFERENCE ONLY

ACP EXPORT DIVERSIFICATION: JAMAICA, KENYA AND ETHIOPIA

Christoplter Stevens

Siismtl» €1 OOl researeh pceseiitecl im preNminary form lor dliisiissloii and eriileaf eominoni

Complete list of Working Papers available at October 1990

15: Macroeconomic Stabilisation, Income Distribution and Poverty: a Preliminary Survey T Addison and L Demery, 1985, £6.00, ISBN 0 85003 094 3 16: Land Tenure Issues in Irrisation Planning Design and Management in Sub-Saharan Africa Mary Tiffen, 1985, £3.00, ISBN 0 85003 096 X 17: Balance of Payments Adjustment and Developing Countries Tony Killick, 1985. £3.00, ISBN 0 85003 097 8 18: The Third Worid's Import Needs and Potential: A European Perspective, Igor Karmiloff, 1986, £6.00, ISBN 0 85003 101 X 19: Macroeconomic Stabilisation in Jamaica: Hie Lessons of Recent Experience Derick Boyd, 1986, £3.00, ISBN 0 85003 102 8 20: Colombia, 1970-85. Management and Consequences of Two Large External Shoclu Juan Jose Echavarria, 1987, £5.00, ISBN 0 85003 109 5 21: Trade and Financing Strategies: A Case Study of Malaysia Mohamed Arif, 1987, £5.00, ISBN 0 85003 108 7 22: Trade and Financing Strategies for Hiailand in the 1980$. Supote ChurumuiuaOttm, 1987, £5.00, ISBN 0 85003 110 9 24: Industrialisation in Sut>-Sataaran Africa: Country case study: Cameroon Igor Karmiloff, 1988, £3.00. ISBN 0 85003 112 5 25: Industrialisation in Sul>-Saharan Africa: Country case study: Zimbabwe Roger Riddell 1988, £3.00, ISBN 0 85003 113 3 26: Industrialisation in Sub-Saharan Africa: Country case study: Zambia Igor Karmiloff, 1988, £3.00, ISBN 0 85003 114 I 27: European Community Trade Barriers to Tropical Agricultural Products Michael Davenport, 1988, £4.00, ISBN 0 85003 117 6 28: Trade and Financing Strategies for the New NICS: the Peru Case Study Jurgen Schuldt L, 1988, £3.00. ISBN 0 85003 118 4 29: The Control of Money Supply in Developing Countries: China, 1949-1988 Anita Saiuonim, 1989, £3.00, ISBN 0 85003 122 2 30: The EfTectiveness of Monetary Policy in Cote d'lvoire Christopher E Lane, 1990, £3.00. ISBN 0 85003 125 7 31: Economic Development and the Adaptive Economy Tony Killick, 1990, £3.50, ISBN 0 85003 126 5 32: Principles of policy for the Adaptive Economy Tony KilUck, 1990, £3.50, ISBN 0 85003 128 1 33: Exchange Rates and Structural Adjustment Tony Killick, 1990. £3.50, ISBN 0 850ra 128 1, 34: Maricets and Govermneiits in Agricultural and Industrial Adjustment Tony Kittidc. 1990, £3.50, ISBN 0 85003 129 X 35: Financial sector polidcs In the adaptive economy Tony Killidk. 1990, £3.50, ISBN 0 85003 131 1 37: Judgbig Succcas: Erahuitlnf NGO Income-Gcnerathig prcjects Ri^er KiddeU, 1990, £3.50, ISBN 0 85003 133 8 3$; ACP E mort DiverrifkaUon: the case oT Zhnbabwe Roger RiddeU ISBN 0-85003-133-8, 1990, £3.50 39: Monetary PoUcjr fai Kenya, 1967-88 Tony KilUdc and F M Uwega, 1990, £3.50, ISBN 0 85003 135 4 40: ACP Export Diveniflcatian: Jamaica, Kenya and Ethiopia Christopher Stevens ISBN 0-85003-136-2, 1990, £3.50 41: ACP Export DivcnincaUon: the case of Mauritius

' McQueen ISBN 0-85003-137-0, 1990, £3.50 tLAh Econometric Study of Selected MoneUiy Policy Issues 1 Kenya F M Mwega ISBN O-85003-142-7 £3.50

00002813

Overseas Development Institute

WORKING PAPER 40

ACP EXPORT DIVERSIFICATION:

JAMAICA, KENYA AND ETHIOPIA

Christopher Stevens

September 1990

ISBN 085003 136 2

OVERSEAS DEVELOPMENT INSTITUTE Regent's College

Inner Circle, Regent's Park LONDON NW1 4NS

Acknowledgements

ODI Working Papers present in preliminary form work resulting from research undertaken under the auspices of the Institute. Views expressed are those of the authors and do not necessarily reflect the views of ODI. Comments are welcomed and should be addressed directly to the authors.

This Working Paper is one of five country papers prepared as part of a study of export diversification by the African, Caribbean and Pacific (ACP) signatories of the Lom6 Conventions. The study is in two parts. The first involved an analysis of trade statistics to identify which states have begun to export which non-traditional products. This is reported in McQueen and Stevens [1989]. In the second, fieldwork has been undertaken in a select group of states to identify the story behind the statistics. The results of these country investigations are reported in three Working Papers. Besides this one on Jamaica, Kenya and Ethiopia, two others examine the cases of Zimbabwe (Working Paper 38) and Mauritius (Working Paper 41).

The overall project is directed at ODI by Christopher Stevens. We are grateful for financial support from the Leverhulme Trust, but they are not responsible for the views expressed here.

CONTENTS

Page

I. BACKGROUND AND OVERVIEW 1

I. 1 The Purpose of the Research 1 I. 2 The Choice of Case Studies 2

II. JAMAICA 3

II. 1 Non-traditional Exports in the Jamaican Economy 3

II. 2 The Policy Environment 5 II. 2.1 Jamaican government policies 5 II. 2.2 US import policy 8 II. 2.3 The EC 12 II. 2.4 Canada's Caribcan 15

II. 3 The Performance of Qothing Exports 15 II. 4 Other Non-traditional Exports to the EC 19 II. 5 Constraints on Production 20 II. 6 Conclusions on Jamaica 21

n i . KENYA 22

III. 1 The Position of the EC in Kenyan Exports 22 III. l . l Total exports 22 III.1.2 Non-traditional exports 23

111.2 The Policy Environment 24 111.2.1 Kenyan government policies 24 111.2.2 The EC 27

111.3 Horticulture 28 III.3.1 Export performance 28 ni.3.2 Production characteristics 35 111.3.3 Major competitors 36

111.4 Canned Pineapple 39 111.5 Qothing and Leather 40

III.5.1 Clothing 41 IIL5.2 Leather 42

111.6 Conclusions on Kenya 42

Page

IV. ETHIOPU 44

IV. 1 The Macroeconomic Framework 44 rv.2 Relative Importance of Non-traditional Exports 44

IV.2.1 nothing 46 rv.2.2 Fruit and Vegetables 51

IV.3 Conclusion on Ethiopia 53

V. GENERAL CONCLUSIONS 54

REFERENCES 57

LIST OF TABLES AND FIGURES

Tables P9«e

1. Manufacturing in the Jamaican economy 4 2. Jamaica: economic indicators 4 3. The JamaicaAJSA Bilateral Textile Trade Agreement 11 4. Jamaican clothing exports, by destination 16 5. Major Jamaican apparel exporters, 1987 18 6. Lom6 rV preferences for fresh (or chilled) fruit and vegetables 29 7. Kenyan exports of fresh horticultural produce, 1973-1988 31 8. Geographical distribution of Kenyan fresh fruit and vegetable exports,

1988/89, by volume 33 9. Geographical distribution of Kenyan green bean exports,

1988/89, by volume 33 10. Geographical distribution of Kenyan cut flower exports,

1988/89, by volume 33 11. Seasonal distribution of Kenyan exports of all fresh fruit

and vegetables, green beans and cut flowers, 1988/89 34 12. EC imports of selected Kenyan non-traditional products 37 13. Sources of supply to the UK horticulture market 38 14. Exports of canned pineapple, 1971-88 40 15. Share of selected non-traditional exports to EC in total exports,

1980-86 45 16. Value of industrial production in constant 1978/79 prices 47 17. Value and costs of production in Ethiopian public and private

clothing and textile industry, 1985/86 47 18. Employment in Ethiopian textiles and clothing, 1985/86 49 19. Domestic sales and exports of textiles and clothing, 1985/86 49 20. Exports of clothing to the EC, 1981-87 50 21. Cost of domestic and imported raw materials for clothing

and textile industry, 1985/86 50 22. Exports of flowers and fresh vegetables to the EC, 1981-87 52

Figures Page

1. Jamaican clothiing exports to USA and EC, 1979-88 16 2. Jamaican clotliing exports, by origin and type 17 3. The share of Kenyan exports destined for the EC, 1973-87 22 4. 'Non-traditional' and total Kenyan exports to the EC, 1976-87 23 5. Manufacturing exports and imports 25 6. Share in merchandise exports of coffee, tea and horticulture,

1983-88 31 7. Growth of 'non-traditional' and total exports to the EC

in current value terms, 1976-87 46

LIST OF ACRONYMS

ACP African, Caribbean and Pacific CAP Common Agricultural Policy Caricom The Caribbean Community

CBI Caribbean Basin Initiative CMT 'Cut, Make and Trim' DCL Designated Consultation Level EC European Community EFVME Ethiopian Fruit and Vegetable Marketing Enterprise EIEA Export Industry Encouragement Act GAL Guaranteed Access Level GDP Gross Domestic Product HCDA Horticultural Crops Development Authority IMF International Monetary Fund JAMPRO Jamaica Promotions Limited JIDC Jamaica Industrial Development Corporation

JNEC Jamaica National Export Corporation JNIP Jamaica National Investment Promotion Limited KETO Kenya External Trade Organisation MFA Multifibre Arrangement

NIP National Indicative Programme

NTC National Textile Corporation

SL Specific Limitation

USAID United States Agency for International Development

VER Voluntary Export Restraint

I. BACKGROUND AND OVERVIEW

1.1 The Purpose of the Research

The 1980s was an inauspicious decade for the trade of the African, Caribbean and Pacific (ACP) signatories of the Lom6 Conventions; in fact, for many of them it was little short of disastrous. As a group, their share of European Community (EC) imports fell sharply and their economic situation worsened dramatically. Clearly, whatever the merits of Lom6 II (1980-85) and Lom6 III (1985-1990) they were no match for the negative economic forces that bore down on many ACP states.

So sharp has been the contrast between the ACP's position as most preferred and yet least successful trading partners with the EC that commentators have begun to question whether Lom6 is worth the candle and, moreover, to assert that since the ACP have gained httle from preferences their potential losses from liberalisation will be correspondingly modest. However, it is too easy to write off the Lom6 trade preferences as valueless; much more cautious conclusions are required. The ACP's poor overall performance reflects two factors:

• first, the Lom6 Convention provides the ACP with either zero or very limited preference over their major competitors for the greater part (by value) of their exports;

• second, ACP exports are more heavily concentrated than are those of other less developed countries on commodities for which world demand is growing slowly.

The evidence of the 1980s is that a number of ACP states have begun to break out of this unsatisfactory export product range, that the provisions of the Lom6 Conventions have faciUtated such diversification, that there is potential for further development during the 1990s, but, finally, that much remains to be done under Lom6 IV to facilitate the process. TTie number of states that have achieved significant diversification is still a minority within the ACP group, but it is not an insignificant one. The success of Mauritius in developing its clothing exports is well-known, but it is by no means alone. In Kenya, for example, horticultural exports to the EC have developed very rapidly over the last five years and are now the country's third largest merchandise export. A recent analysis of non-traditional ACP exports has identified 28 states as significant exporters and, although many have exported only a small number of new products, one-third have exported six or more, and five have exported more than fifteen [McQueen and Stevens, 1989]. Moreover, diversification is not limited to the more advanced ACP states, such as Kenya, Mauritius, Zimbabwe and C6te d'lvoire. It also includes poorer countries at lower levels of economic development, such as Ethiopia, Sudan and Ghana.

The analysis of trade statistics indicates which countries have begun to export which products, but it does not explain why this has happened, nor does it indicate whether there are residual restrictions in the EC market that are preventing an even more rapid diversification. It is the object of the country case studies to deal with these two neglected areas.

1.2 The Choice of Case Studies

The ACP is a very diverse group of states. This diversity extends beyond wide differences in the level of economic development and embraces geographical location, language and culture. Hence, no small group of case studies can be representative of the group as a whole.

Despite this, it is possible to identify a manageable nimiber of countries which, while not fully representative, do illustrate some of the main characteristics that recur frequently within the group. The five countries given particular attention in this project not only feature prominently in the trade analysis but also cover several of the main ACP sub-groups.

The contrasts provided by Jamaica, Kenya and Ethiopia are particularly interesting. Jamaica, a middle-income country with a relatively well-developed manufacturing base, is, perhaps, one of the countries that might have been expected to benefit most from the Lom6 preferences. Yet its exports of non-traditional goods to the EC have been surprisingly low imtil very recently. Kenya, too, might have been expected to be a significant exporter of manufactures to Europe, but its diversification has been primarily in the agricultural sector. The level of economic development in both cotmtries stands in stark contrast to that of Ethiopia, the poorest ACP state in terms of GNP per capita. The level of Ethiopian exports of non-traditional products to Europe is much lower than that of the others, but it illustrates that diversification is not the prerogative of middle-income states. Moreover, Ethiopia, unlike either Jamaica or Kenya, has managed to diversify into both manufactured and agricultural products.

Between them, these three states illustrate the differing importance of the various Lom6 trade preferences in relation both to each other and to those offered Ijy the ACP's other trading partners. They show also the interaction of demand-side constraints (such as the rules of origin) and those on the supply-side (notably tmsupportive government policies) and the scope for aid to ease the bottlenecks that Umit further diversification. In none of the cases do the non-traditional exports 'solve' the problem of stagnant demand for traditional exports. The new markets into which they have diversified are highly competitive; diversification is a continuing exercise, not a once-for-all shift. But they confirm the findings of the statistical analysis, that the ACP are not somehow incapable of diversification. Furthermore, they provide some evidence that the Lom6 preferences have made a contribution.

II. JAMAICA

II.l Non-traditional Exports in the Jamaican Economy

Non-traditional exports have a smaller share in the exports of Jamaica to the EC than of the other countries being studied, and are of very recent origin. This is for three reasons. First, two of the main traditional exports to the EC - bananas and sugar -are covered by special regimes under which prices have held up much better throughout the 1980s than has been the case with the ACP's other traditional exports. Second, although there has been a very rapid growth of non-traditional exports from Jamaica, it has until very recently been directed almost exclusively at the USA. Third, and most important, government policy changed away from import substitution to favour export growth only in the early 1980s.

The dominance of the USA as a market for Jamaica's non-traditional exports is a primary focus for this country study. What are the reasons for the US emphasis? Is it due solely to the fact that the USA is a very large market on Jamaica's doorstep that has absorbed all of the output that Jamaica has managed to produce? Or have there been especially favourable characteristics of US trade policy, or negative features of EC policy?

Jamaica's non-traditional exports, at least outside the Caricom area, are almost exclusively of manufactures, notably garments. Processed agricultural or temperate agricultural exports (apart from rum) are still at a fledgling stage. Manufacturing's share of GDP has fluctuated at around 16%, but its share of exports doubled between 1982 and 1987, reaching 20% by 1988 (Table 1). This growth contrasts sharply with the overall performance of exports during this period. The volume of total exports declined steadily to 1985, and although it then increased the rise was of only 12% in the three years to 1988 (Table 2).

Jamaican statistics do not allow a precise calculation of the share of manufacturing output, or employment, accounted for by exports because, of course, some companies are involved in both domestic and foreign markets. However, it is possible to identify the impact of those manufacturers operating within the export free zones established during the second part of the 1980s (see below). Since these companies account for a very large part of the growth in manufactured exports of recent years, figures on the free zones provide a broadly accurate picture of the overall impact of export-oriented manufacturing. Net foreign exchange earnings from the free zones increased from US$ 1.1 million in 1982 to a peak of USS 31.6 million in 1987, with a slight downturn thereafter. By 1987 these earnings were equivalent to some 10% of the trade deficit. Employment in the free zones, also largely absorbed with export-oriented manufacturing, has likewise grown rapidly from virtually nil in the early 1980s to 11,400 by 1987, again with a dip in the following year. In early 1987 it was reported that over 18,000 people, mainly women, were employed in clothing either in the free zones or in the rest of Jamaica [JNEC, 1987].

Table 1: Manufacturinii in the Jamaican economy

mi 1982 1983 1984 1985 1986 1987<f>198^'>

Share of Manufacturing in GDP'''^ (%) 15.0 16.0 16.0 16.0 16.0 17.0 17.0 16.0

Growth of Nfanufiicturing''^ <%) 1.1 6.7 1.9 4.2 0.5 3.7 5.2 -0.9

Share of Exports'''^ (%) 10.0 9.0 9.0 16.0 16.0 20.0 20.0

Gross Bqmrts &om Free Zones (USSmn) 14.3 14.3 19.0 43.5 69.7 -

Net Foreign Bxdiange earnings from Free Zones (USSmn) 1.1 1.8 2.4 7.1 10.8 31.6 27.2

Employment in Free Zones ("000) 0.9 1.0 3.1 5.1 7.8 11.4 7.1

Notes: (a) GDP at constant prices; (b) Annual rate of growth at constant prices; (c) Manufiactured goods + machine & transport equipment + miscellaneous

manu&unures;

(p) preliminary; (e) estimated.

Source: The Planning Institute of Jamaica. Economic & Social Survey Jamaica rxineston, 1989), various issues.

Table 2: Jamaica: economic Indicators

1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988

Volume of exportsf"-' 180.7 159.0 174.2 177.5 173.9 176.0 173.0 1263 123.9 127.1 100.0 105.6 112.9 1123

Trade balance'*^ -161 -135 71 81 -64 -76 -323 -442 -439 -335 -436 -248 -362

Exchange rate - USjW 0.9 0.9 0.9 1.4 1.8 1.8 1.8 1.8 1.9 3.9 5.6 5.5 5.5 55

Exchange rate - Ecu'°> - 1.8 25 2J 10 1.8 1.7 3.1 43 5.4 6.4 65

Cbttsumer prices' '' 17 19 21 28 37 46 52 56 62 80 100 115 123 133

Notes: fa^ 1985 = 100. (b) USS millions; minus sign = deficit. (c) JS per USS/Ecu. (<i) 1985 = 100.

Source: IMF. International Financial Statistics.

II.2 The Policy Environment

II.2.1 Jamaican government policies

[a] Macroeconomic policies

During the 1960s and most of the 1970s the emphasis of government policy was on import substitution. Through a combination of tariffs and quotas on the one hand and an over-valued exchange rate on the other, the government both increased the attractions of the domestic market for local manufacturing and reduced its capacity to compete on world markets. Although Jamaica has always been a highly trade-dependent economy (with an average ratio of trade to GNP of over 100% throughout the 1970s), exports outside the Caribbean region were almost exclusively based on sugar, bauxite/alumina and bananas. As the external balance worsened during the 1970s, import licensing proliferated. By 1980 370 items were covered by licensing, compared to only 60 in 1970 [Harrigan, 1989]. The volume of exports has declined steadily since Lomd I was signed (Table 2). Until 1984 the nominal exchange rate changed hardly at all vis-d-vis the US dollar, even though the armual rate of inflation was well into double digits.

From 1977 onwards a number of attempts have been made to reorient the economy by whichever government was in power. Export stimulation has been a declared objective of policy for most of this period although actual policies, moulded tmder pressure from resource constraints and/or political demands, have not always been so supportive. Moreover, policy changes (such as devaluation and the removal of import controls) have often failed to overcome the expectations of businessmen sufBciently for them to reorient their attention from the domestic to the export market.

Since the mid-1980s the export-oriented policies have paid off, at least in respect of foreign capital. For domestic manufacturers the net effect of government policies has been the stimulus to exports invoked in rolling down protection and devaluing balanced against the constraints associated with severe foreign exchange shortages and sharp measures to restrict demand, such as high interest rates. Following an agreement with the IMF in 1987, for example, the government reduced tariffs affecting the manufacturing sector. Tariffs on raw materials were reduced from 16% to 10% over one year, and those on capital goods were reduced to 20% over four years. In 1988, capital goods imports were also relieved of consimiption duty (which was formerly set at 15% ad valorem), as well as excise duty and sales tax where applicable. Stamp duty was also reduced (from 30% to 20%), and there were duty concessions on 'high use' capital goods, Le. those quickly consumed in the production process such as sewing machine needles [Steele, 1988:116]. At the same time, however, interest rates are extremely high (31% at the time of research), and the capacity to import inputs and capital equipment is restricted by the shortages of foreign exchange and the effect of devaluation.

[b] Fiscal incentives to export

The government has established two regimes for companies engaged exclusively in exports, together with special arrangements for those engaged in supplying both the domestic and export market. These provisions have had to be dovetailed into the requirements of Caricom.

Under the Export Industry Encouragement Act (EIEA) 1956, a company producing exclusively for export outside Jamaica and Caricom may be declared an 'approved export manufacturer' in relation to 'an approved eq>ort product'. Such firms, which may be located anywhere on the island, are exempt from the payment of income tax on profits for ten years and, until 1987, could elect either to import duty free a range of inputs or to claim a tax rebate on exports outside Caricom. In 1987, as part of the agreement with the IMF, a twin-track approach was introduced. New companies establishing under the EIEA are now given the option of either accepting the pre-existing concessions or of paying full import duty on raw materials but recovering a proportion of this under the Export Rebate Scheme whilst continuing to benefit from the tax exemptions. The tax rebate on exports is currently set at the rate of 7.5%. The IMF reasoning was that the existing scheme was an inefficient way of promoting exports and that a rebate on tax paid would be more effective.

More recently a number of export free zones have been created under the Jamaica Export Free Zone Act of 1988. As with the EIEA, beneficiary companies must export their entire output. There are currently four principal free zones: two in Kingston, one in Montego Bay (which has recently become host to a teleport high speed data transmission service) and one in Hayes. In addition, a number of large companies have been designated export free zones in their own right, e.g. the Tropicana ethanol plant. Companies within the free zone rent their infrastructure from the government on advantageous terms (made possible because the free zones have been constructed using the profits from Kingston port and concessional financing from the World Bank) and their imports and exports outside Caricom are entirely free of tax and duty. The companies also enjoy total exemption from income tax in perpetuity.

The principal difference between the two systems concerns the handling of foreign exchange. Companies in the free zones must pay for all their local costs in foreign exchange, but there are no controls on the repatriation of profits. Companies approved under the EIEA, by contrast, can pay for inputs in local currency but require permission to repatriate profits. For these reasons, the free zones are particularly attractive to those foreign companies that have established in Jamaica to undertake outward processing in the garments sector for re-export to the USA. All of the companies in free zones are foreign-owned or joint ventiu-es in which the foreign partner is dominant, while those established under the EIEA are a mixture of foreign and indigenous.

[c] Institutional support

Export promotion is currently undertaken by a single parastatal, Jamaica Promotions (JAMPRO) Limited, formed in 1988 from the merger of three

agencies dating from the early 1980s with overlapping responsibility in the field: Jamaica National Investment Promotion Limited (JNIP), the Jamaica National Export Corporation (JNEC) and the Jamaica Industrial Development Corporation (JIDC). It provides a range of services including investment promotion (such as assistance in obtaining permits, identifying local investment partners, locating land and factory space), and trade promotion (including certification, market intelligence, and training).

JAMPRO has been particularly active in encouraging export-oriented clothing manufacture. It is reported that US industry sources consider the Jamaican programme to be the most sophisticated of those mounted by the Caribbean suppliers [Steele, 1988:110]. Among the facilities it provides, JAMPRO has retained the services of Kurt Salmon Associates to provide management consultant support for larger firms. For smaller companies, it works in partnership with USAID to develop management and operator training programmes. A more general training function is provided by the Garmex Academy under the goverrmient's Heart youth training programme, which is funded by a special payroll levy. These technical assistance programmes are not without their critics, but they appear to have had a significant impact building up the country's competitiveness as a garment producer and exporter.

Under its trade agreements with the USA, Jamaica has export quotas for specific clothing items. The existence of these quotas has provided the government with a tool for encouraging the clothing firms operating in the country to develop in preferred directions. The global quotas for each product are distributed between clothing manufacturers by JAMPRO. Whilst the main criterion for determining each company's quota is its historical level of supply, a number of additional criteria are also applied in order to encourage certain areas of activity. These additional criteria are:

• diversifying exports to non-US markets; • for those producers in the free zones, sub-contracting to companies

outside the zones; • giving higher local added value in employment; • including non-restricted items in the export mix; • the use of US capital and other material and equipment; • location outside the Kingston corporate area.

Hence, the govenunent has deliberately used its allocation of the US quotas inter alia to encourage the geographical diversification of clothing exports to the EC.

Other aspects of the institutional framework for exporting are less supportive. Export administration is highly complex. The Third National Export Plan of 1987-89 even described it as a disincentive to exporting in its own right [cited in Steele, 1988:114]. Partly because of the demands made by the authorities in the countries with which Jamaica has preferential trade agreements (USA,

Canada and the EC), there are a large number of forms which have to be completed and hurdles that have to be overcome. Progress along this course is hindered because of a lack of automaticity in the system and a relatively high degree of administrative discretion.

Registering as an exporter with the Trade Board is fairly straightforward, but to obtain a certificate of Jamaican origin for exports is more difficult, involving two steps. First, all exporters of textile products have to be registered as being capable of producing the goods they wish to export, and stringent criteria are applied in this test. Second, each individual shipment has to be certified as being of Jamaican origin (except in the case of 807 shipments - explained below - which are specifically exempted because of the limited processing undergone in Jamaica). For exports to the USA, a special visa and certification system has been established to ascertain the appropriate import regime to be applied. To obtain many of the exporters' incentives made available by the goverimient, exporters have to obtain a Bank of Jamaica compliance certificate, verifying that export earnings have been sold to the Bank at prevailing rates of exchange, and tax compliance certificates.

Additionally, customs clearance tends to be more onerous than in many countries because of the problem of drugs. Jamaica has been adversely affected by the stringent penalties now imposed by US authorities on import consignments in which drugs are found to be concealed. In May 1988 US customs imposed a fine of USS 65 million on Evergreen Shipping Line after finding ganja in three sealed containers from the Jamaica Biscuit Company. The result was that Evergreen discontinued its Caribbean service with effect from July 1988. This was a severe blow since it was the largest carrier on the Jamaica-USA route, handling 75% of the containers shipped in 1987. Following negotiations with the government, the decision to pull out of Jamaica completely was reversed but Evergreen resumed only on a scaled-down service. Later in the year, the Jamaican government introduced stringent measures designed to curb the export of drugs. All containers are now stripped before loading. When first introduced, this caused considerable logistical problems for exporters. Not only did the inspection cause delays but, more seriously, it introduced a significant element of uncertainty into logistical plaiming. It appears that the system has now been improved so that it is a hardship for exporters but not a major handicap.

II.2.2 US import policy

[a] General provisions

o 807

The growth of Jamaican clothing exports to the US is undoubtedly the result of US trade policy. The principal instrument has been Item 807 of the Tariff Schedules of the United States Annotated (TSUSA), as amended by various measures tmder the Caribbean Basin Initiative (CBI). 807 encourages outward processing by US companies. It is thus akin to

the provisions in the EC's bilateral agreements with the Southern Mediterranean countries, but is much more complex and directive.

Although 807 applies to a range of manufacturing sectors, it is garment-making that has been the prime beneficiary. Some of the largest and most efficient US clothing manufacturers and distributors have demonstrated interest in the scheme, suggesting that it is genuinely attractive to the US industry. 807 shipments have come to account for a significant proportion of US clothing imports: up to 10% in 1987 [Steele, 1988:5].

Under the standard 807 procedure, when US manufacturers assemble goods overseas from components manufactured in the USA and then reimport the goods, they do not have to pay import duty on the value of the US content of the finished product. The benefits of 807 are purely fiscal; the provision does not formally ease the quantitative restrictions imposed by the USA on clothing and textile imports under the Multifibre Arrangement (MFA). In practice, however, countries supplying under 807 are less likely to have a quota imposed on a particular product or, if one is imposed, it is likely to be larger than would otherwise apply. To benefit from 807 the cloth must have been cut within the USA, but not necessarily from US-produced fabric.

• 'Super 807'

The 807 provision has been amended in several respects for Caribbean countries falling under the CBI. As originally announced in 1983 the CBI would have covered textile products, but these were then excluded following a dispute within the US administration between the Departments of State and Commerce. To fill the gap, the US finally approved a CBI Textile Program, which was announced by President Reagan in February 1986. This invited CBI states to enter into special bilateral textile trade agreements with the USA that would provide additional market access opportunities. At the same time, the USA tightened its quota restriction on imports from the Far East, thus increasing the incentive for Asian exporters to relocate to a quota-free country.

Countries establishing bilateral textile agreements with the USA under the CBI Textile Program became eligible for what is now known as 'Super 807'. For the textile industry the significant feature of Super 807 is that it increases the US content of the final product. As under 807, the role of the host country is limited to the assembly of pre-cut pieces but, to benefit from Super 807, the pieces must have been cut from US fabric (defined to include wholly-US material and fabrics woven or knitted in the USA from imported yam). For the host country. Super 807 offers some relief fi-om MFA quotas. For goods falling under the Super 807 rules of origin, the USA has negotiated guaranteed access levels (GALs) which, though formally a quota, are set at a level at which they will not normally be binding.

At the same time, however, restrictions on goods falling foul of the Super 807 rules of origin were tightened. The object was to encourage manufacturers to export under Super 807, Le. using US-made fabric. Goods not qualifying for GAL treatment were subject to either designated consultation levels (DCLs) or specific limitations (SLs), for which quotas were set at much lower levels than the correspoiuling GALs.

Early implementation of Super 807 was hindered by major bureaucratic problems, together with a difficulty in sourcing some of the required materials within the USA. More fundamentally, the Caribbean states complained that the provisions made it impossible for participating countries to build up substantial clothing industries which were not simply offshore assembly operations. This fundamental criticism would have been addressed to a certain extent by the provisions of the Caribbean Basin Economic Recovery Expansion Bill. But this bill had a difficult passage to the statute book. When it was passed finally, in August 1990, it excluded preferential treatment for textiles and clothing.

• 7Vv«n plants

One further item of US trade poUcy with importance for the Jamaican clothing industry is Section 936 of the US Internal Revenue Code. As a resuh of negotiations between the US govenmient and Puerto Rico over the use of this section, Puerto Rico offers low-cost loans firom 936 deposits in the island's banks to finance the establishment of so-called 'twin plants' in CBI countries. A number of such plants have been established in Jamaica, although Costa Rica and the Dominican Republic are more preferred locations, partly because of language.

[b] Agreements with Jamaica

Jamaica was the first Caribbean country to respond to the invitation by the USA to enter into bilateral textile programmes, although it is not the largest regional supplier under 807 (the Dominican Republic and Mexico hold the top two positions). Under the agreement Of August 1986 three groups of products were established: group 1 (GALs for Super 807); group 2 (DCLs for non-Super 807 products); and group 3 (for other products not subject to restriction but liable to be 'called' under the market disruption provisions of the agreement). The items placed in these three categories, and the quotas to which they were subject, have been revised three times - in March 1987, September 1987 and in the Montego Amendment of April 1988. The details of the agreement and its subsequent amendments are given in Table 3.

In most cases the GALs have been non-binding: only the knitted shirts/blouses and hosiery quotas were exceeded in 1988 [Steele, 1988:129]. These agreements apply only to simple assembly and sewing of pre-cut pieces, not to more complex processes such as 'Cut, Make and Trim' (CMT). US import controls on CMT goods are more severe. Nonetheless, even on CMT the US is by no means significantly less liberal than is the EC. Although quotas do

Table 3: The Jamaica/USA Bibiteral Textile Tnuie Agreement

A. Apparel categories placed under restraint in the USA/Jamaica bilateral trade i

Category Prvduct Group Access leveb 1.9.86-31.1Z87 1988-89

('000 dozen) (mn sye) ('000 dozen) (mn sye)

338/339/638/639 Knitted shirts and blouses, cotton and man-made fibre

I II

850 6.1 575 3.6

695 495

5.0 3.6

347/348/647/648 Trousers etc., male and female, cotton and man-made fibre

I n

1547 25.5 695 12.4

1250 653

223 11.6

331/631 Cotton and man-made fibre gloves I n

1500 5.3 300 1.1

1320 350

4.6 1.2

349/649 Colton and man-made fibre brassieres I 2575 12.4 2200 10.6

340/640 Woven shirts, cotton and man-made fibre

Vam-dyed sub-categoty II 375 9.0

275 6.6 450 300

10.8 7.2

B. Further appani categories placed under restraint ID the USA/Jamaica bilateral trade agreement, March 1987

Category Product Group Annual levels 1.6.87-31.1187 1988-89

('000 dozen) (mn sye) ('000 dozen) (mn sye)

341/641 Woven blouses, cotton and man-made fibre I 117 1.7 200 19

345/845 Sweaters, cotton and other vegetable fibre I 29 1.1 50 1.8

352/652 Underwear, cotton and man-made fibre I II

554 74 58 0.8

950 100

128 1.4

632 Man-made fibre bcsiety I II

729 3.4 58 0.3

1250 100

5.8 05

340/640 Woven shirts, cotton and man-made fibre I 200 4.8 200 4.8

C Further categories phiced under restraint in the USA/Jamaica bilateral trade agreement, September 1987

Category Product Origmal DCL NewSL 1988 1988 1989

('000 dozen) (mn sye) ('000 dozen) (mn sye) ('000 dozen) (mn sye)

338/339/638/639 Knitted shirts 495 3.6 575 4.2 610 4.4 340/640 Woven men's shirts 450 10.8 325 7.8 345 8.3

Yarn dyed sub-limit 300 7.2 275 6.6 291 7.0 continued /...

Table 3 (... / continued)

D.

1. The agreement was extended to the end of 1992.

Z The trouser DCL (347/348/647/648) of 653,000 dnrni for 1988 and 1989 was converted to an SL of 750,000 dozen (11.6 mn j ^ ) with 6% a year gttjwth rate and standard flexibility thereafter.

3. The knitted shirt and blouse SL (338/339/638/639) was increased from 575,000 dozen to 695,000 dozen in 1988 with a growth provision of 6%.

4. The underwear DCL (352/652) was increased from 100,000 dozen to 300,000 dozen (4.1 mn sye) for the lifetime of the agreement.

5. A number of new DCLs were established from July 1, 1988. The level for the six months until the end of the year was merely prorated for subsequent years with no growth provision as it was envisaged that the categories would begin to qualify for 807a treatment by the end of 1988. The DCLs, on an annualised basis, were:

336/636 (dresses), 98,000 dozen (4.4 mn sye); 349/649 (brassieres), 250,000 dozen (1.2 mn sye); 447 (men's woollen trousers), 5,000 dozen (0.1 mn sye); 342/642 (skirts), 175,000 dozen (3.1 mn sye) (125,000 dozen in fust six months).

6. The following GALs were increased with effect from January 1, 1989:

352/652 (underwear) from 950,000 to 1,550,000 dozen (20.9 mn sye); 340/640 (woven shirts) firom 200,000 to 300,000 dozen (7.2 mn sye); 341/641 (blouses) from 200,000 to 375,000 dozen (5.4 mn sye); 347/348/647/648 (trousers) from 1,250,000 dozen to 2 mn dozen (3.6 mn sye); 352/652 (underwear) from 950,000 dozen to 2 mn dozen (27.0 mn sye).

7. The following additional Gals were established with effect from 1989:

336/636 (dresses), 125,000 dozen (5.7 mn sye); 342/642 (skirts), 200,000 dozen (3.6 mn sye); 447 (woollen trousers), 12,000 dozen (0.2 mn sye).

Source: Steele, 1988, pp-lU-lM.

exist for CMT goods they have not been binding in most cases, there has been a problem only with sweaters.

II.2J The EC

Jamaica is an original signatory of the Lom6 Convention and, hence, has benefited, at least in theory, from duty-free access to the European market for the past 15 years. Evidently, this has not resulted in a surge of exports! Part of the reason for this is to be found in government policies during the 1970s that tended to discourage exports. And part is also due to movements in exchange rates. Trade with the EC is particularly vulnerable to exchange rate movements. Not only is it affected by Jamaican government policies that increase the real effective exchange rate of the Jamaican dollar, but it is also

influenced strongly by the US doUarrEcu relationship. Even when the government is pursuing a 'realistic' exchange rate policy, the foreign currency with which the Jamaican dollar is aligned is the US dollar. Given the high proportion of Jamaica's trade that takes place with the USA this is inevitable. However, one consequence is that when the US dollar appreciates against the Ecu, Jamaican exports cannot avoid becoming more expensive in terms of the European currencies. During the first part of the 1980s, of course, the US dollar appreciated very substantially against the Ecu, as did the Jamaican dollar (Table 2). It is therefore not surprising to find that Jamaican exports of manufactures to Europe did not take off.

In addition to these 'supply side' problems, however, exporting to the EC faces one major 'demand side' obstacle: the rules of origin. Their impact on clothing exports has been particularly noticeable. The requirement in the rules of origin that woven clothing be produced either from yam or from cloth imported from the EC has effectively prevented the development of woven clothing exports to Europe. Jamaica has one textile mill, but it does not produce a quality of cloth suitable for export clothing. Because of Jamaica's distance from Europe and, no doubt, the dominance of US and Far Eastern companies, the import of cloth from the EC is considered financially unviable [JAMPRO, 1989a]. The EC rales of origin are more stringent than those appMed by either the USA or Canada. Whereas Super 807 does require Jamaican exporters to use US-made fabric, it is complemented by standard 807 treatment which can use non-originating materials.

The rales of origin do not inhibit only clothing exports. Two biscuit manufacturers have attempted to export to the EC under the Lom6 Convention. Neither has succeeded in obtaining duty-free access, although one is able to continue to export to the ethnic market in the UK even when it has to pay duties.

The two companies have faced different problems. The first uses Jamaican domestically-milled flour in all of its biscuit production. The Lom6 rales of origin specify that biscuits qualify for dufy-free treatment only if the wheat from which the flour is milled is produced either in the ACP or the EC.

Under Lom6 III biscuits were included in List A of Annexe II, which meant that more than one process had to be undertaken in an ACP state to qualify under the rales of origin. The standard test for whether a product originates in an ACP state or not is that the Customs Tariff Heading Number under which it is classified is different from the numbers applying to any of its component parts. However, for the products covered in List A, this 'change of tariff heading' criterion is insufficient. The list specifies the additional processing that has to be undertaken in order to confer originating status. In the case of biscuits, it is specified that the process of manufacture fi-om flour is not of itself sufficient, and, hence, the country of origin of the raw materials determines whether or not the final product benefits fi-om the Lom6 preference. Jamaica produces no wheat. All of the wheat supplied to its mills is in the form of US food aid under PL 480. This does not prevent the biscuits receiving preferential access to the Canadian market under Caribcan, even though Canada is in a position

exactly analogous to that of the EC of wishing to encourage its own wheat exports rather than those of the USA. The company considers it infeasible to obtain an EC-originating raw material. This would require it to import flour on its own account solely for the biscuits intended for the U K market. This is considered to be both impractical and financially unviable.

The second company is willing to purchase European flour. Its normal poMcy is to import the flour for its export biscuits rather than to purchase locally because the PL 480 wheat is of variable quality. However, its usual source is Canada rather than the EC. Nonetheless, it considers that it is financially viable to buy European flour for the purposes of manufacturing biscuits for export to the UK.

A first shipment of these biscuits was made in January 1989, in the belief that they qualified for duty-free access under the rules of origin. On arrival in the UK the customs imposed a 35% levy on the grounds that the product did not qualify for Lom6 treatment. EC import duties on biscuits are in two parts -a fixed component and a variable component. The Lom6 Convention provides for the fixed component to be waived on imports of all types of biscuits from the ACP. But the variable component is waived only on certain biscuits, primarily those with a high starch content (over 50% in some cases and over 65% in others). The biscuit in question was a sweet biscuit with a starch content of only 45.66%. The imposition of the levy made the transaction financially unviable, the company lost money, and no further exports to the EC are likely unless either an appeal against the levy is eventually successful or the rules are changed.

Provisions exist under the Lom6 Convention for 'derogations' (Le. temporary exemptions) from the Rules of Origin. Under Lom6 III such exemptions could be granted for a period of three years in the case of middle-income countries like Jamaica, with a possibility of renewals for a maximum period of two years. Neither of the biscuit manufacturers has sought recourse to the derogation procedure, which has been widely criticised because the EC has tended to interpret the rules narrowly and their administration has been slow and costly. Few derogations have been sought anywhere in the ACP, although all the applications that have been made have eventually been granted - albeit with certain amendments. From the evidence available it would appear that neither biscuit manufacturer had a particularly strong case for a derogation. This is because of the essential temporary nature of the waiver: applicants must show that they will be able to fulfil the normal rules of origin after the end of the transition period. The problem for Jamaica is that the logistics of flour supply require a permanent, not a temporary, waiver.

Under homi IV the derogation procedures have been improved, in terms both of their administration and the criteria for approval. Of potential importance to Jamaica is that derogations will now be approved automatically if there is a minimimi level of value added in an ACP state of 45%, provided that this does not cause 'serious injury' to an established Community industry. Since no applications have yet been made under the new derogation procedure it is

premature to assess the extent to which the amendments of Lome IV will benefit Jamaica. However, this is clearly an area for scrutiny.

II.2.4 Canada's Caribcan

In addition to its preferential access to the US market under the CBI/807, and the EC market under the Lom6 Convention, Jamaica, along with the other Commonwealth territories in the Caribbean, has signed with Canada a special trade agreement known as Caribcan. Under it, Jamaica enjoys duty-free access for its exports into the Canadian market, apart from a number of excluded products, provided that rules of origin are met. The rules of origin specify that a minimum of 60% of the ex-factory price of the goods must originate in Jamaica, in other Caribcan beneficiaries, or in Canada. This value added threshold is not as high as might appear at first sight since the ex-factory price includes overheads and a reasonable level of profit. Unfortunately for the Jamaican garment industry, the exclusions from Caribcan include both textiles and clothing, together with footwear, luggage, handbags, leather garments, lubricating oils and methanol. These products benefit only from the generalised system of preferences, if at all.

113 The Performance of Clothiny Exports

The story of clothing exports is essentially one of the 1980s. In the 1970s, import substitution pohcies, combined with an overvalued exchange rate and increasing economic dislocation, kept exports outside the Caribbean area to a low level.

The growth of exports has been largely synonymous with the takeoff of exports to the USA from 1984. Total clothing exports increased from USS 5.1 million in 1979 to USS 11.3 million in 1983; they then increased very rapidly in the following years to reach USS 221.7 million by 1988 (Table 4).

Until very recently the growth was due almost exclusively to exports to the USA, but a start has now been made in exporting to the EC. After several years of virtually zero sales, exports to the EC increased to USS 2.8 million in 1987 and then USS 10.1 million in 1988 (Figure 1). By 1988 the EC accounted for 5% of Jamaica's clothing exports, making it the only significant market after the USA (which held a commanding 93% of the total). Partial figures for 1989 show that the growth of exports to the EC has continued.

The increase in exports is also closely associated with the free zones, although exports from the Jamaican customs territory have also increased. In 1982 the free zones accounted for only 3% of Jamaican clothing exports. By 1988 this had increased to 50% (Figure 2).

There has also been a shift in the structure of production. The share of CMT in total exports has risen slowly. In 1986 Jamaica exported only USS 40 million of CMT goods; by 1988 this had risen to USS 91 million, giving CMT 44% of the total. About half of the major firms are now involved in some CMT activities (Table 5).

TaMe 4: Jamaican clothing exports, by destination (USS million)

%ofim Market 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 total

Caticom 2.4 Z4 3.4 5.4 4.6 2.6 0.7 0.4 0.8 1.2 0.5 USA 2.2 3.4 2.8 l U 6.1 26.1 29.6 973 179.8 2053 93 EC 0.7 0.4 0.4 0.01 0.2 0 0 03 18 10.1 5 Canada 0.3 0.1 0.2 0.04 0.02 0.2 0.2 0.5 23 23 1 Other 0.5 1.1 0.5 0.6 0.6 0.4 0.4 0 1.2 2.8 1

TOTAL 5.1 7.0 7.1 173 113 32.6 31.1 98.6 186.9 221.7 100

Sources: JNEC, Annual Report on Garments 1985 (March 1986).

JNEC, The Third National Export Plan for Jamaica 1987-89 (January 1987).

J A M P R O , Annual Report on Garments Sector 1988 (March 1989.

Fignre 1: Jamaican clothing exports to USA and EC, 1979-88 (USS million)

220

1979 1981 1983 1985 1987 1980 1982 1984 1986 1988

Sources: As for Table 4.

Figure 2: Jamaican clothing exports, by origin and type

120 I

19S2 1983 1984 1965 19B6 1987 1988

Source: JAMPRO.

The increased share of CMT is associated with the development of exports to the EC. CMT exports to the USA are Mmited since the products fall outside Super 807 and may be subject to quotas. In the case of the EC, however, there is no such problem. Because of the rules of origin, exports to the EC are concentrated in knitted goods, notably sweaters and T-shirts. Exports to the USA, by contrast, are much more broadly based; in 1987 the principal items were: trousers, slacks and shorts, bras, shirts/blouses and coats/jackets [JAMPRO, 1989a].

How far has this growth of exports benefited the economy? Like other outward processing promotion schemes, 807 and Super 807 assume that there is scope for 'enlightened self-interest' by encouraging domestic manufacturers to undertake part of their operations in lower-cost third countries. The US industry is assumed to benefit because, by lowering its total production costs, it is better able to compete with cheap imports. And the third party host is also assumed to benefit through the establishment of labour-absorbing, export-oriented manufacturing and the vehicle that this provides for US technical, managerial and marketing skills.

The criticism most often levelled at outward-processing is that it creates a dependent industry m the host that becomes familiar with only part of the production process (assembly and sewing), gaining little experience of design, cutting, marketing, and that operates on very low profit margins at a high level of vulnerability to sudden changes in demand over which it has no control. In short, the industry creates (possibly temporary) employment but little else.

Table 5: Major Jamaican apparel exporters'"'', 1987

Company Country of ownership

Parish of location^'''

Type of operation Product line

Afuia Kniltere (Jamaica) Ltd Hong Kong KEFZ CMT Sweaters Akom Ud USA MBFZ 807 Sweatsuits Apparel Handlere Ltd Jamaica St James 807 Sweatsuits Bagutta Garment (Jamaica)

Lid Hong Kong KEFZ CMT Pullovers, sweatsuits and shoru Budget Dress Shop Ltd USA Kingston CMT Dresses Clawtc Manufacturing Ltd Jamaica St James 807 Trousers/Macks Danah Bra Company I-td USA St Mary 807 Bras, panties Davon Corporation Ltd Jamaica Kingston 807/CMT Ttonsei^acks, thins East Ocean Textiles Ltd Hong Kong KEFZ CMT Shirts, blouses, trousers/slacks and

tkiru Garment People Lid (The) Jamaica St James CMT/807 Shorts, blouses Gazapati Jamaica Ltd UK, India, USA KEFZ CMT Blouses, shorts Hofab Manufacturing Company

Ltd Jamaica Kingston 807 Pantyhose Jacon Industries (Montego)

Ltd Jamaica St James 807 T-shins Jamaica Needlecraft Ltd USA KEFZ 807 Brassieres Jamtei; Ltd USA KEFZ 807 Trousers Jockey International

(Jamaica) Ltd USA Hanover 807 Underwear Jog Togs Ud USA Kingston 807 Children's outerwear and sleeimear Just Brands Assoc. of

Jamaica Ltd USA KEFZ CMT Trousers, shirts, jackets and skirts Karia Lucea Ltd USA Hanover 807 Bras, panties L & Q Apparel Ltd Jamaica Kingston CMT/807 Shirts, blouses, shorts and skirts La Mods Manufacturing Ltd Jamaica Kingston 807 Ladies' outerwear Lawrence Manufacturing

Company Ltd USA KEFZ CMT Shirts Magma Enterprise Ltd South Korea/

Canada KEFZ CMT Noel of Jamaica Ltd USA Kingston CMT/807 Trousers, shifts and skirts Regal Foundations LJd Jamaica St Catherine 807 Bras and girdles Regent Exports Ltd Jamaica Kingston 807 Trousers, shirts Satisfaction Sewn Products

Ltd Jamaica Kingston 807/CMT Trousers, slacks, shorts, shirts and

btlouses Sinotcx Jamaica Ltd Hong Kong KEFZ CMT Shorts, trousers, slacks and skirts Strathleven Ud Jamaica St James 807 T-shirts United Fashions Company Ltd South iCorea St Catherine CMT Blouses, skirts, dresses and shirts Universal Knitwear Ud Hong Kong Kingston CMT Sweateis, shins and skins Williamson-Dickie Jamaica Ltd USA MBFZ 807 TrousenAlacks Windmill Garment

Manufacturers Ud Jamaica Trelawny CMT/B07 Trousers, slacks, shins Yoffi Industries Ud USA St Catherine 807 Bras and panties

Notes: fa1 Coratianies which shiDoeti apparel valued at more than USS 1 million in 1987. (b) KEFZ = Kingston Export Free Zone; MBFZ = Montego Bay Free Zone. (c) CMT = Cut, make and trim.

Source: Steele, 1988.

The strength of this criticism can be reduced if the host industry develops from simple assembly to more complex operations, such as CMT. In 1987 it was estimated that the import content of '807' exports was as high as 85% of the f.o.b. value, whilst for CMT the figure was lower, at 40-60% [JNEC, 1987]. Although few of the US companies that have established in the Caribbean region under 807 have moved into CMT, there is a major exception in Jamaica, where one large American company has done so. Moreover, four Hong Kong-based firms signed an agreement with JNEC in 1985 both to diversify into Europe and to increase CMT exports [Steele, 1988, p.52 and Table 16.1, and JNEC, 1986]. In 1983 virtually all Jamaica's clothing exports to the USA were of simple assembled articles; by 1988 CMT accounted for 48% of export value. And exports to the EC are exclusively CMT.

II.4 Other Non-traditional Exports to the EC

There are not, as yet, any major non-traditional exports other than garments to the EC, even though a number of attempts have been made to initiate them and there do exist some other non-traditional exports to the USA. The problems that the two biscuit companies have run into with the EC's rules of origin are instructive.

Among the non-traditional goods that Jamaica has begun to export to North America, but not yet to the EC, are ornamental plants and furniture. By 1987 exports of ornamental plants had reached US$ 3.8 million, of which two-thirds were to USA and one-quarter to Canada; exports to the EC were only USS 200,000, or 5% of the total. There are twenty-six companies established in the commercial ornamental nursery business, with 523 acres. The main plants are heUconia, ornamental foliage and anthurium, plus small areas of orchids, chrysanthemums, roses and other flowers. About 7,000 people are employed in the industry [JAMPRO, 1989b].

The attractions of the North American markets are related to both policy and geography. Ornamental plants receive favourable duty concessions under the CBI. Moreover, Jamaica is close to the major horticultural centre of Miami. Finally, there is a market demand for the type of product Jamaica can grow.

The main constraint on accessing the EC market is distance. JAMPRO's view is that the volume of production must increase before it makes commercial sense to attempt to export to Europe on a significant scale. One company, which exports primarily to USA and, to a lesser extent, to Canada, did export to the U K during the 1970s. But it had two imfortunate experiences with its commercial associate there. This led it to suspend exports to the U K and it has been wary of restarting.

The export furniture business involves some 35 firms employing 3,500, largely skilled labour, and concentrating on reproduction antique wooden furniture for the top end of the US and Bermudan markets. This appears to be a sensible market orientation and there appears little reason to develop exports to the EC. The North American market is the best one for high-priced reproduction furniture, and the obvious development strategy for the Jamaican industry is to increase its capacity to supply that market [UNILES, 1987].

II.5 Constraints on Production

There are well over 100 clothing manufacturers registered as exporters (Table 5). They fall into three broad groups. Group 1 consists of about 60 relatively large companies which compete successfully on the international market. Some are mainly 807 operators, while others have CMT activities as well. In 1987, 10 of the 65 CMT producers accounted for some 85% of such exports, while 10 companies were responsible for a similar share of 807 exports [JNEC, 1987]. They run factories with over 100 machines and employ between 150 and 3,000 workers. The majority are foreign-owned and have marketing organisations in their main market. The main source of foreign capital is the USA, followed by Hong Kong and South Korea. In group 2 come about 40 companies which are mainly locally-owned and operated. In the past they tended to concentrate on the domestic or Caricom markets. However, the combination of government enthusiasm and declining Caricom demand has led many of them to become 807 contractors and a few also to handle CMT orders. Very few have marketing links in their major markets. They have between 60 and 99 machines and employ between 60 and 188 workers. Many experienced financial difficulties in 1988 and some closed. Finally, there is in group 3 a nucleus of small factories with under 60 machines.

The main constraints facing the clothing sector are financial. They arise because a significant number of companies have accumulated debts which are now unsustainable given, on the one hand, extremely high rates of interest and, on the other, the volatility of the 807 market and its very narrow profit margins. A review of the clothing industry's problems by Coopers and Lybrand in 1988 identified the main problem for those companies in difficulties as being debt consolidation and a cut in interest rates [Coopers and Lybrand, 1988]. At the present time interest rates stand at 31% and have been high for some years as part of successive governments' demand control policies designed to accompany structural adjustment.

Other problems concern the labour force. The difficulty is not one of nominal remuneration rates: Jamaican wage rates are low relative to those of the USA. A recent analysis by Kurt Salmon Associates for JAMPRO suggests that savings over US manufacturing costs (excluding materials) for a range of products are between 55% and 65% for 807 production and between 30% and 38% for CMT [JAMPRO, n.d.]. Prevailing Jamaican wages average just 13% of those in the USA.

The problem is rather one of the quality of labour: companies claim that there is room for improvement on this score. The Coopers and Lybrand report lists poor training, absenteeism and theft as subsidiary problems identified by the companies it surveyed. There appears to be widespread criticism by companies of the training provided by the Garmex Institute.

Although organised labour traditionally has been politically powerful in Jamaica, the export-oriented clothing industry is relatively lightly unionised. There have been some labour disputes involving Asian companies operating in the free zone. In the mid-1980s five major trade unions wrote to the government complaining that pay and conditions in the export-oriented industries were poor, albeit in line with minimum statutory wage rates [Long, 1986]. Their complaint was that much overtime is required and is low paid; they claim that employees are expected to work a minimum

of 12 hours per day, six days per week, and that refusal to undertake overtime leads to dismissal.

11.6 Conclusions on Jamaica

After a long period of economic difficulties, Jamaica has achieved an impressive growth of non-traditional exports. Doubts have been expressed over the sustainability of this process in the light of rapid changes in the US clothing industry and the domestic problems of high interest rates, foreign exchange shortage, etc.. Nonetheless, in terms of the questions posed by the present study, Jamaica provides a clear example of an ACP state that has capacity to export competitively the kind of manufactured goods that the Lom6 Convention was intended to stimulate.

The question that then arises is why exports to the EC have been so feeble. In part, the explanation must be that the US has been able to absorb most of the goods that Jamaica has been able to produce. Nonetheless, it is also clear that part of the explanation Mes in the difference between US policy and that of the EC. Whereas the US government has positively encouraged a partial shift in the international division of labour, the EC has been much more restrictive. The Lom6 rules of origin have prevented Jamaica exporting to Europe the kind of clothing that has been most prominent in exports to the US and have failed to provide any preference to the two recent attempts to export non-traditional goods other than clothing. This failiu-e is the more distressing because both USA and Canada have provided more sympathetic treatment to Jamaica than has been meted out imder the 'special relationship' of Lom6. Under 807 the US has given preferential treatment to woven clothing made from non-US fabric. Likewise, Canada has granted preferences on goods containing non-Canadian raw materials, as in the case of biscuits made from US wheat.

m . K E N Y A

111.1 The Position of the EC in Kenyan Exports

IIL1.1 Total exports

The EC has been a major export market for Kenya since before Lomi I was signed (Figure 3). Its share of Kenyan exports slumped to about one-third in the early 1980s, but for most of the period it has been between 40% and 50% of the total. Of the European Community national markets, the British is the most important with around one-fifth of the total. Germany is the only other large market, with the remaining member states each taking less than 10% of Kenyan exports.

Figure 3: The share of Kenyan exports destined for the EC 1973-1987 (%)

ss 50

45

40

35

30

25

20

15

10

5

° 1973 1975 1977 1979 1991 1983 1985 1987 1974 1976 1978 19BQ 1992 1984 19aB 1988*

Note: • Provisional

Source: Central Bureau of Statistics, Economic Survey, various issues.

After the EC, Kenya's principal markets in 1987 were Uganda (8%), USA (6%) and Pakistan (5%) [Kenya 1988, Table 53]. The relative unimportance of USA is in contrast to the experience of Jamaica and some of the other ACP states that have diversified successfully, such as Mauritius. The US share in Kenyan exports rose during the 1980s - from 3% in 1980 to a high of 9% in 1986, although it then fell sharply the following year. However, this has still put it behind UK, West Germany and, in most years, the Netherlands.

ni.1.2 Non-traditional exports

The share of 'non-traditional' commodities in Kenya's exports to the EC has increased substantially during the period since Lora6 I was signed (Figure 4). The term 'non-traditional' is used in the sense in which it has been applied in the larger research project of which this study of Kenya forms a part. That is, it refers to a broadly selected group of some 179 commodities (mainly at the 6-digit level) which does not include any of the well-established, dominant commodities such as lightly-processed coffee, cocoa or minerals. As in any such selection for a whole group, there are certain anomalies at a country level. In the case of Kenya, for example, 'non-traditional' exports include canned pineapple, which some might argue has been a feature of trade sufficiently long for it now to be regarded as a traditional export.

Figure 4: 'Non-traditionai' and total Kenyan exports to the EC 1976-87 (Ecu million)

1976 1 378 1990 19B2 1384 19BB 1977 1979 1981 1983 1985 1987

Source: Eurostat.

Bearing in mind these caveats, non-traditional Kenyan exports to the EC grew three times as fast as total exports between 1976 and 1987. In consequence their share (by value) of total exports rose from 10% in 1976 to 23% in 1987.

There was also a broadening of the range of non-traditional commodities. In 1976 the bulk of such exports were various kinds of processed finiit (mainly canned pineapple) and lightly-processed hides and skins. By 1987, by contrast, a range of horticultural products and a small amount of clothing had joined the

list. As a result of this rapid growth, horticulture is now Kenya's fourth largest merchandise export, being exceeded only by coffee, tea, petroleum and, on the non-merchandise account, tourism.

Commodity diversification has proceeded further in Kenya's exports to the EC than to other destinations. Non-traditional exports to the USA in 1987, for example, were at a very low level (equivalent to only some 3% of the level of exports to the EC) and were concentrated on leather goods and woven clothing. The relatively low share of the USA in Kenya's non-traditional exports (which contrasts with the experience of other ACP states such as Mauritius and Jamaica) may be due to the very limited role of manufactures. Unlike many of the other ACP diversifiers, Kenyan non-traditional exports are heavily concentrated on agricultural rather than manufactured goods. Whereas the EC offers preferences under the Lome Convention for both agricultural and manufactured goods, this is less true of other industrialised countries.

This commodity diversification has been particularly helpful because Kenya's traditional exports are characterised both by considerable instability and by declining prices relative to imports in recent years. The Kenyan economy is a very open one, with the combined value of imports and exports equivalent to 49% of GDP in 1988 [Killick and Mwega, 1990:8]. Worid prices for coffee and tea have been very volatile with, for example, coffee experiencing a 39% rise in the realised price between 1985 and 1986 and a 37% fall in the following year. Over the period since the mid-1970s, however, the general trend of prices for traditional exports in relation to the prices of imports has been downward. The index of purchasing power of exports has declined from 114 in the period 1974-78 to 86 by 1984-88 [Killick and Mwega, 1990:Table 1].

m.2 The Policy Environment

III.2.1 Kenyan government policies

[a] Macroeconomic policv

In practice, if not in precept, industrial policy has tended to put import substitution before export diversification. This applies particularly to manufacturing, which may explain why this sector has played only a minor role in the commodity diversification of exports to the EC. This absence is the more surprising since Kenya has a well-developed manufacturing sector and was a major regional exporter during the colonial and post-colonial periods. During the period of Lom6 I (1975-80) real manufacturing GDP grew by an annual average of 7.2% (higher than the average for the total GDP, which was 5%), and in the period of Lom6 II it increased by 4% (as against 3.2% for total GDP) [Lewis and Sharpley, 1990:Table 7.2].

Government policy statements during the 1970s and early 1980s were couched in terms of the need to reduce protection to domestic sectors. An Export Compensation Scheme was introduced in 1974 to assist manufacturing exporters with cash subsidies intended to offset the protective effects of tariffs on inputs

and the cascading effects of domestic excise and sales taxes [ibid.:!]. Practice, however, has tended not to conform with this precept. Since 1971 there has been a consistent rise in the average level of protection as import licensing has become an increasingly important element of both macroeconomic management and the system of protection for domestic manufacturing industries. This was particularly marked during the 1970s, when a range of quantitative import restrictions were introduced. During the 1980s the protective effect has been moderated somewhat with the replacement of quantitative controls by tariffs.

The most important source of manufacturing grovrth over the past two decades has been the increased domestic demand. One analysis of the sector claims that two-thirds of output growth during the twenty years to 1984 was accounted for by domestic demand, and just over one-quarter from import substitution (Figure 5). Export growth contributed only 5% to the increase.

Source: Lewis and Sliarpley, 1990.

The one exception to the bias in recent years has been in relation to the exchange rate. Diversification has undoubtedly benefited from devaluation. The period of rapid growth in horticultural exports has coincided with a gradual, but substantial, devaluation of the Kenya Shilling, which has been subject to a real depreciation of the order of 28% between 1985 and 1988 [Killick and Mwega, 1990].

[b] Fiscal incentives to export

The fiscal regime tends to act as a disincentive rather than a stimulus to export diversification. Provision does exist for a 20% tax rebate for exports, but this applies only to manufactures which, as noted above, are notable mainly by their absence. The rebate is considered by the Ministry of Finance to be compensation for taxes paid on the imported capital goods required to produce the export rather than as a subsidy. It is for this reason that it has not so far been available for horticulture exports with few clearly identifiable imported inputs.

Two particular aspects of the fiscal regime appear to have caused problems. These concern taxes on aviation fuel and duties on imported packaging materials. The major constraint on further expansion of exports at the present time appears to be air freight capacity. Unlike Ethiopia (see below), there is no shortage of south-bound flights to bring freight costs down to an acceptable level. But there is not always sufficient space on scheduled airlines to absorb supply and there appear to be constraints in hiring charters. A widespread view, expressed by both private sector and Government sources during the fieldwork for this project, is that aviation fuel prices in Kenya have been higher than in neighbouring countries as a result of Government tax policies [see, for example. Chamber of Commerce, 1989:23]. It is alleged that this has made it difficult to attract charter flights. The Government recently reduced the tax on aviation fuel considerably; it will be important to monitor the effect on air transport availability.

Another constraining factor, which appears Hkely to become increasingly important, is the absence of adequate packaging materials [Chamber of Commerce, 1989:24]. In an effort to increase the unit value of exports, some exporters are attempting to move into the export of pre-packed vegetables which are suitable to be placed on supermarket shelves without further repackaging. These have two advantages: the level of value added in Kenya is increased, and the journey from farm to shop is reduced by one day - increasing the freshness of the product on the shelves and, hence, its price. Government policy is to provide trade protection for the domestic packaging industry. Packaging materials for bulk exports are supplied by a single company, which does not produce the type of material required for pre-packs. At present, therefore, the packs have to be imported, and face an import duty of 45% plus a sales tax of 17% (even though they are re-exported unmediately). This reduces the financial competitiveness of the resulting export.

Such duties are not the only taxes applied to export production. The local authorities in the main horticultural growing areas (Machakos, Mem and Malindi) impose a production cess [Chamber of Commerce, 1989:29]. Also a 15% cess is levied on imported hides and skins required for leather exports.

[c] Institutional support

The horticultural industry is supervised by the Horticultural Crops Development Authority (HCDA), set up in 1968. This registers all exporters and provides

technical and marketing information to both exporters and producers, for which it levies a charge of 10 cents/kg. In conjunction with the Ministry of Agriculture and Livestock Development, the HCDA's technical services department provides production advice to farmers. It also has a marketing department which undertakes limited marketing for very small farmers and provides market intelligence to the industry as a whole. As part of this, it is a member of Coleacp, an organisation set up in 1973 under the aegis of the EC Conmiission, which offers a specialist professional framework for cooperation between ACP exporters and European importers of tropical fruits, vegetables, flowers and spices. HCDA is also responsible, in collaboration with the Central Bank of Kenya, for monitoring prices and foreign exchange remittances mto Kenya.

However, the horticulture trade appears to be a highly individualistic affair, with exporters undertaking much of their own marketing. Partly because of this, a frequently voiced complaint concerns problems in ensuring payment by importers. At present there appears to be no system for obtaining references on potential importers or of checking the prices they actually receive on goods sent on commission {Le. to be sold at the best price obtainable).

Pre-export inspection services are provided by the Ministry of Agriculture. This operates a unit at Nairobi Airport which provides phytosanitary certificates and other documentation required to gain entry to the EC.

There is a Kenya External Trade Organisation (KETO) within the Ministry of Commerce and Industry. This provides general government support for horticultural, as for other, exports. Together with HCDA it participates in trade fairs, giving special emphasis to horticultural products, and it has conducted contact promotion programmes and sent trade missions abroad for horticultural marketing. Although most contacts are made by the exporters themselves, the KETO claims some credit for the development of markets for some horticuhural products, such as mushrooms, and for the development of handicraft product markets.

The govenunent is considering establishing an export processing zone, with the idea of developing clothing exports particularly in mind. However, studies are still being imdertaken on the practical implementation of this. There already exists provision for manufacturing under bond within a factory, although this does not seem to have figured in exports to the EC.

There also exists a National Chamber of Commerce which has some interest in promoting non-traditional exports. It recently sponsored a mission of horticultural exporters to the main EC markets.

IIL2.2 The EC

The Lomd Convention trade preferences of most relevance given Kenya's actual performance in diversification are the, often tightly delineated, concessions for products falling under the Common Agricultural Policy (CAP). The CAP

regime for horticultural products is complex. The basic rule is that the system for supporting European farmers is relatively lightly structured, without the mandatory intervention buying and variable import levies characterising the cereals and meat regimes. For the fresh products of most interest to Kenya, the normal regime applying to imports is that the EC levies an ad valorem tariff and also establishes a 'reference price'. Countries exporting to the EC are obliged to sell their goods at a 'minimum import price' equal to the reference price plus the tariffs. Failure to comply results in a countervailing levy being imposed to bring the cost of imports up to the required level. Hence, it is possible to export fresh fruit and vegetables to the EC, but only if the landed price exceeds the level at which domestic produce is sold.

For the ACP, and some other third party suppliers, concessions are made on CAP products. The concessions take the form of full or partial rebates of the ad valorem tariff. But there are two provisos. The first is that ACP suppliers must still respect minimum import prices. In other words, they are unable to undercut domestic European produce but they retain a larger share of the proceeds from any exports they do make. This helps them, of course, to compete with other third party suppliers that have to pay the full tariff. The second proviso is that these concessions are Umited to a fked quota for some products. These quotas may be very small, e.g. the quota in Lom6 IV for small winter cucumber is 100 tonnes for the whole of the ACP group! Table 6 provides a list of the preferences that will be accorded to horticultural products in Lom6 IV.

In addition, the Lome Convention provides for duty-free access to the EC market for manufactured goods that fulfil the rules of origin. Although Kenya has not taken great advantage of these concessions in the past, there is a possibility that it might do so in the future.

Kenyan exports have not been subject to any formal restrictions (other than those specified in Lom6). But surveillance and 'voluntary export restraints' (VERs) on cut flowers have been proposed (although not implemented) at various times by the EC Commission, and by the Netherlands and German governments.

Horticulture

IIIJ.l Export performance

[a] Share in total exports

Kenya has exported horticultural goods to the EC since before the first Lom6 Convention (Table 7). However, the trade was at a fairly modest level until the mid-1980s. The big jump in exports took place in 1984, when the voltmie increased more than threefold over the previous year; it was from 1984 that the govertmient began to devalue the real effective exchange rate.

Table 6: homi IV preferences for fresh (or chilled) Ihilt and vegeUbles

Regime Proilucl(s) Quota (tonnes)

Exemption bom customs duties without marketing timetat>le celery other than celeriac, sweet peppers,

courgettes, other vegetables, pistachioe, pecans, other nuts, grapefruit, other citrus fruit, limes, melons, pawpaws, fruit of the

fniil

None

Reduction of the duly by 60% Tomatoes (other than cheny tomatoes), from 15 November to 30 April

2,000 tonnes

Progressive abolition of customs duties

Cherry tomatoes, from 15 November to 30 April

2,000 tonnes

Progressive abolition of the customs duty Onions, from 1 Febniaiy to 15 May 800 tonnes

Progressive abolition of the customs duty Gariic, fit>m 1 February to 31 May 500 tonnes

Progressive abolition of the customs duty from 1 November to 31 December

Chinese cabbage 1,000 tonnes

Progressive aboUtion of the customs duty from 1 July to 31 October

Iceberg salad 1,000 tonnes

Progressive abolition of the customs duty from 1 Januaiy to 31 March

Carrou 800 tonnes

Progressive abolition of the customs duty Horse-radish

Progressive abolition of the customs duly Salad beetroot 100 tonnes

Progressive aboUtion of the customs duty Small winter cucumbeis 100 tonnes

Progressive abolition of the customs duty from 1 October to 31 December

Anicholies 1,000 tonnes

> Progressive abolition of the customs duty from IS August to 15 Januaiy • 40% reduction from 16 Januaiy to 31 January

) ) ) )

Asparagus

Progressive abolition of the customs duty other mushrooms

Progressive abolition of the customs duty Walnuu 70O tonnes

Progressive abolition of the customs duty from 1 November to 30 April

Figs (fresh) 200 tonnes

• Progressive abolition of the customs duty from 15 May to 30 September - Above this quantity, and throughout the year, an 80% reduction of the custonu duly

) ) ) )

Oranges 25,000 tonnes

• Progressive abolition of the customs duty from 15 May to 30 September • Above this quantity, and throughout the year, an 80% reduction of the customs duty

) ) ) )

Mandarins and other similar citrus hybrids 4,000 tonnes

Progressive reduction of the customs duty by 50%

Apples 1,000 tonnes

Continued / ...

Table 6 (... / continued)

Regime Product(s) Quota (tonnes)

Progressive reductioil of ttie customs duly Dy50%

Peats 1,000 tonnes

Progressive atjotiUon of tlie customs duty from 1 September to 30 April

Apricou 2,000 tonnes

Progressive reduction of the customs duly from 1 November to 31 March

Cherries 2,000 tonnes

Progressive reduction of the customs duly from 1 December to 31 March

Peaches 2,000 tonnes

Progressive reduction of the customs duty from 15 December to 31 March

Plums 2,000 umnes

Progressive abolition of the customs duty Sloni SOOUmnes

Progressive abolition of the customs duty from 1 November to end February

Strawberries 1,500 tonnes

Progrcsshw abolition of the customs duty Mixtures exclusively of dried nuts (0801 and 0802)

Reduction of the common customs tariff to:

• 3% for fruit of the species Vaccinium macrocarpon and Vaccinium

. 5% for other fruits of the Vaccinium species

Since 1984 there has been substantial growth in the volume, value and unit value of exports. There was an average annual growth of 10% in export volume over the five years to 1988. In value terms, exports grew in ciurent prices from below K f 10 million in 1979 to K£95 million by 1988. The unit value of Kenya's horticultural exports has grown more rapidly than the average for all exports. Between 1984 and 1988 it rose in current terms by an annual average of 4.6%; by contrast, the price index of all exports rose an annual average of only 0.7%. Horticulture export unit values have risen consistentiy throughout the period covered by Table 7. Indeed, the effect of high CAP price levels may be discerned in the more than 100% jump between 1974 and 1975.

As a result of both the increase in volume and the faster-than-average growth of unit value, horticulture's share of total merchandise exports has increased steadily. From only 3% in 1983, it had reached 10% of the total by 1987, a position which it maintained in 1988. Coffee's share in 1988, by contrast, stood at 27%, with that of tea at 20% (Figure 6).

Table 7: Kenyan exports of firesh horticultural produce, t973-1988'''>

Vobime Value Unit value Price index Share in total ofttllexforls merchandise exports

(•000 tonnes) (K£ million) (K£ltonne) (1976=100) (%)

1973 10.2 1.5 147 46 0.7 1974 11.3 1.7 150 69 0.5 1975 13.1 4.2 321 86 1 1976 18.0 5.1 283 100 1 1977 18.8 6.4 340 108 1 1978 21.0 8.0 381 114 1 1979 21.3 9.7 455 128 2 1980 22.3 11.4 511 154 2 1981 23.3 12.6 541 170 2 1982 24.6 13.6 553 186 2 1983 28.9 17.5 606 223 3 1984 103.7 54.2 523 268 7 1985 84.5 53.0 627 264 7 1986 110.4 66.1 599 281 7 1987 136.9 77.1 563 240 10 1988W 151.5 94.8 626 276 10

Notes: (a) Prior to 1973 a different, non-comparable, system of valuation used; (b) Provisional.

Source: Central Bureau of Statistics, Economic Survey, various issues.

Figure 6: Share in merchandise exports of coffee, tea and horticulture. 1983-1988 (%)

45 45

40 coffee

/ \ B 35 / \ Taa

30 1

^^^-""'^ \ H o r t i c u i t .

25

20 ^ X ^ ^ , ^ ^ - ^ ^ a -15

10 ^ ' 5

Q 1983 1984 1985 1986 1387 1988

Source: Central Bureau of Statistics Statistical Abstract 1988. Economic Survey 1988.

[b] Geographical distribution

In volume terms, exports of fresh fruit and vegetables are distributed reasonably evenly across the EC (Table 8). The UK and Italy are the largest markets (each with 27% by volume of 1988 exports), followed by the Netherlands, France and Germany. There are marked seasonal variations in the relative importance of the national markets; UK dominance is much stronger during the European summer months than the average.

These figures, however, do not reflect accurately the relative value of the national markets. Rather, they reflect the type of products that dominate in each. Exports to the UK tend to be higher volimie/lower value items, and include a large proportion of 'Asian vegetables'. These were an early focus of Kenyan horticultural exports but are now relatively less important in terms of overall value as the sector has diversified into higher value items. Currently the most important of these are green beans and cut flowers, which combine high unit values with a substantial export volume. France and UK are the principal markets for green beans (Table 9), whilst Netheriands and West Germany take the bulk of cut flowers (Table 10).

The importance of France as a market for green beans is more marked than appears from Table 9. This is because a significant part of the share going to Belgium during the European summer months is probably re-exported to the French market. This arises because France has imposed a ban on imports during the summer months in order to give preference to francophone African exporters. Initially, the ban applied for the whole of June-September. In 1986 the Government of Kenya tabled a formal complaint at the EC-AC? Ministerial Council, and in 1988 the French Government agreed to permit a quota of 150 toimes in each of June and September, thus limiting the total ban to just two months. However, it is believed that a significant indirect traffic, particularly via Belgium, takes place to evade these restrictions. Under Lome IV the ban will be removed altogether.

In the case of cut flowers, the dominance of the Netherlands is to be expected since most of Kenya's exports are sold on the Dutch flower auctions. Hence, of course, their ultimate country of destination could be anywhere in the worltj (including the other EC states). The concentration on one national market, therefore, does not imply an inability to compete in the other markets: rather it reflects the structure of the industry. Moreover, the existence of substantial exports direct to West Germany suggests that Kenya is well able to compete in unrestricted markets. Nonetheless, there is some evidence that Kenya's range of markets is more limited than it might be. Sales to UK, for example, tend to be low partly because flowers at Covent Garden are sold in the carton. Good quality packaging is therefore an important part of the product, but the Kenyan packaging is not sufficiently competitive. In the Dutch auctions, by contrast, this problem does not arise because the flowers are sold in buckets and subsequently repackaged.

Among the lesser value items France and Britain tend to be the most important markets, although this is not always the case. For avocadoes, France is by far

Table 8: Geographical distribution at Kenywi fresh fruit and vegetable exports. 1988/89, by volume (%)

1988 1989 1988 Market Market

Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Total

U K 26 23 22 26 34 33 34 35 54 59 63 47 27 Netherlands 15 13 26 31 20 20 15 6 6 3 6 11 12 France 16 10 12 14 16 21 27 36 7 5 4 9 12 W. Germany 15 12 12 16 15 13 10 7 7 11 13 16 10 Italy 16 32 15 2 2 - 4 4 6 3 - 1 27 Belgium 3 2 5 3 3 2 3 3 10 9 7 11 3 Other 9 8 8 8 10 11 7 9 10 10 7 5 9

Source: H C D A .

Table 9: Geographical distribution of Kenyan green bean exports, 1988/89, by volume (%)

1988 19S9 1988 Market Market

Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Total

U K 14 31 25 29 31 38 34 28 50 60 62 48 35 Netherlands 3 3 4 6 3 2 3 2 2 3 3 3 3 France 62 45 46 49 48 42 46 58 12 5 6 13 37 W. Germany 5 5 5 3 5 6 5 3 4 5 6 4 5 Italy Belgium 12 10 16 8 9 7 8 6 27 25 22 31 15 Other 4 6 4 5 4 5 4 3 5 2 1 1 5

Source: H C D A .

Table 10: Geographical distribution of Kenyan cut flower exports. 1988/89, hy volume (%)

1988 1989 1988 Market Market

Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep Total

U K 7 7 6 4 7 6 6 2 - 1 21 19 6 Netherlands 45 45 64 64 53 65 68 55 78 26 26 30 53 France 2 3 1 10 2 2 4 - - - 2 1 2 W. Germany 41 40 25 29 33 24 17 34 12 62 47 41 32 Italy Belgium Other 5 4 4 3 5 3 5 8 10 11 4 8 7

Source: HCDA.

the largest market, followed by UK and Germany. In the case of fresh pineapples, too, France is the main importer, but in this case Germany is second and Great Britain third. The Netherlands appears at first sight to be a substantial importer (taking 19,300 tonnes in 1988), but over 60% of this was subsequently re-exported [Qiamber of Commerce, 1989:8]. U K and France are the main importers of mangoes. Once again, the Netherlands features prominently in the list but re-exports a significant share to other EC states. A similar position applies to passion fruit, with the same three major importers and Netherlands, once again, re-exporting a significant share. U K is the only country in Europe with a consistent demand for okra. Germany is the leading importer of capsicum.

[c] Seasonal distribution

There is a marked seasonal bias to Kenyan exports to Europe. Exports during the European winter months are higher than during the summer, and for the more specialised, higher value products (such as green beans and cut flowers) this bias is particularly marked (Table 11). Almost two-thirds of cut flower exports by volume occur in just four months, and in the case of green beans the four most active months accounted for a little under half of the total.

Table 11: Seasonal distribution of Kenyan exports of all fresh fruit and vegetables, green beans and cut flowers, 1988/89

1988 1989

Oct hfov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

Green beans (tonnes) 1003 1125 1405 1459 1210 1063 1394 1825 881 812 617 983 % of annual total 7 8 10 11 9 8 10 13 6 6 4 7

Cut flowers (tonnes) 1355 1555 2289 2552 1563 1276 765 262 114 72 239 763 % of annual total 11 12 18 20 12 10 6 2 1 0.6 2 6

All horticulture ('000 tonnes) 4.4 6.3 6.0 5.6 4.4 4.6 4.0 3.8 Z5 2.3 ZO 3.0 % of annual total 9 13 12 12 9 9 8 8 5 5 4 6

Note: Percentages do not add up due to rounding.

Source: HCDA.

In value terms, the winter bias is even more marked. This is apparent from EC trade statistics which classify imports during the period October to June in a different category from those from July to September. In 1987, the latest

year for which data have been obtained, Kenya's exports of green beans during October to June totalled Ecu 14.2 million, or an average of Ecu 1.6 million per month; in the three summer months, by contrast, exports totalled only Ecu 2.3 million, or a monthly average of Ecu 0.8 million. Even these figures may understate the seasonal bias because exports are possible only in some of the winter months. In the case of capsicum, for example, there is a shortage of supply in the EC market only from January to April.

The reason is obvious: competition is much more severe during the European summer fi:om domestic produce, so that prices fall and the incentive to exp)ort is reduced. It is only fairly recently than Kenya has tried to export during the simimer months. The main stimulus has come from the producers, who seek an all-year source of income. Their demands are strong because they are mainly to be found in the coffee-growing area, where their alternative crop is an all-year source of income and employment.

IIIJJ Production characteristics

There are no detailed statistics on the horticulture sub-sector. For example, there are no official data on the number of people employed in horticulture production, but it has been variously estimated as tens of thousands. Flower cultivation alone is said to employ about 7,000 people. One reason for the lack of data is that the growth of horticulture appears to have passed more or less unnoticed by the Government. The annual Statistical Abstract, for example, does not even identify horticulture as a separate agricultural sub-sector, let alone as a major export! This may account for the fact that the sector has benefited fi-om no special Government support or subsidies.

Over one hundred exporters are registered with the HCDA, but only 40-50 currently are active. A typical exporter may deal with 50-60 farmers. The dividing line between farmers and exporters is somewhat bluned; the largest producers undertake their own exports, and many exporters also have some production capacity.

The farmers range from smallholders on half an acre to large farms with 400 or more acres to horticulture. The relative shares of total production accruing to large farmers and smallholders is unknown. However, it is imderstood that large-scale farms in the area around Naivasha dominate the cut flowers market and that large farms (and the larger smallholders - with holdings of over 5 acres) are also particularly important for high value vegetables where quality control is especially stringent. As Kenya changes its product mix to increase the share of high value items, it will present a challenge to smallholders to maintain their share of the industry.

In addition to air space and packaging, noted above, another constraining factor is the absence of adequate common-user cold storage facilities at Nairobi airport. Some of the large producing and exporting companies have their own cold storage facilities, but the facilities available to small exporters are said to be inadequate. This has two adverse effects. First, it limits the possibility of 'pre-cooling' produce, a process that is favoured by importers in the EC.

Second, it makes the small exporters very vulnerable if cargo space is not available as planned. If, for any reason, cargo space fails to be made available after the produce has reached the airport, the exporter either faces the complete loss of his consignment or must sell at a knockdown price to an exporter with surplus space. Plans are afoot to construct additional storage, but they are still at a preliminary stage.

III3J M^or competitors

[a] ACP

Kenya's penetration of the European market has undoubtedly been facilitated by the tariff concessions provided under the Lom6 Convention. These range from 8% for avocadoes to 26% for canned pineapples. However, Kenya is not alone in enjoying these preferences. Precisely the same preferences are available to all ACP states, a number of which have similar production characteristics to those of Kenya. Some of the sharpest competition is coming, therefore, from other members of the ACP group.

Data is provided in Table 12 for the most important 'new' exports to the EC, showing the relative performance of the ACP in the European market compared to other developing countries. In most cases Kenya is the largest ACP supplier, but it is not the only one for winter green beans, canned pineapples and miscellaneous cut flowers. In most cases the growth in ACP exports to the EC compares quite favourably with that of all developing countries. In the case of cut flowers, for example, ACP exports of miscellaneous fresh cut flowers have increased over twice as rapidly as those from developing countries as a whole for the period 1976 to 1987, despite the fact that the growth has largely been at the end of that period.

[b] Other competitors

The ACP are not the only countries that have been accorded trade preferences on horticultural products by the EC. There is a complex set of agreements, notably vwth the countries of the Meditertanean, which provide similar tariff reductions subject to quotas and calendars. Moreover, some states are able to compete effectively on the EC market even though they receive no preferences.

The principal source of competition varies with the product. On strawberries, for example, Latin America has just entered the EC market, as a result of which prices have become very depressed. Costs of production in Latin America are alleged to be lower and the frequency of transport is greater. Latin America is also a major competitor for mangoes. With avocadoes, Israel (which has a customs union with the EC) is a major competitor at certain times of the year, having been able to extend its growing season. South Africa is also a competitor, but Kenya has a certain seasonal advantage since it is able to begin shipments one month earlier. For green beans, Egypt is a main competitor (together with Tanzania and Rwanda), but Kenya maintains its position as the highest-quality producer. On flowers, Spain is a particularly strong competitor, followed by Israel.

Table 12: EC Imports of selected Kenyan non-tntditlonal exports

Pnxluci (NIMEXE mmber) %

ACPexpomio EC (Ecu mlUiai)

1976 19S7

GrtmOi in apora u> EC (% p.tL) ACP AUldc

Principal ACP Es^omrs (PrACP)

Short in 19^ PrACP aports, %

Haricot beaiia: Ocl.-Junc (070145) 13 5.7 263 15 14 Kenya, Senegal,

Burkina, Ethiopia 2.5

July-Sept. (070147) 17 03 23 20 18 Kenya OS

Avocadoe (08016O) 8 1.0 2.6 10 15 Kenya 05

Pineapples (canned) 20-26 44.2 46.3 4 7 Kenya, Swaziland, C»te d'lvoirc

1.5

Miscellaneous cut flowers; Nov.-May (060369) 17 16.2'*.' 14.1 36 17 Kenya, Ethiopia,

Zimbabwe, Mauritius Swaziland

1.0

June-Oct. (060319) 240 O.iC') 16 23 11 Kenya, Mauritius, 0.1 cote d'hroire, Jamaica, Zimbabwe, Ethiopia

Carnations: Nov.-May (060355) 17 4.4W 9.6 9 8 Kenya 2.0 June-Oct. (060305) 24 O.2W 3.9 33 15 Kenya 0.8

Orchids (060307) 24 0.06fW 0.04 0 0 Kenya 0.01

Notes: W (b) (c)

Common Customs Tariff; 1980; 1977.

Source: Eurostat.

An indication of the range of competitors is provided in Table 13, which shows the main sources of supply to the UK market for those horticultural products of particular interest to Kenya. The number of competith'e countries varies considerably between products, with avocadoes and mangoes being the most crowded. A recent report by representatives of the Kenyan industry claimed that the country has already lost, or is on the verge of losing, a substantia] share of the market for mange touts (lost to Zambia and Zimbabwe), courgettes and capsicum (both lost to Spain and Cyprus), passion fruit (Zimbabwe) and avocadoes (South Africa, Israel and Australia) [Chamber of Commerce, 1989:32]. It also asserted that there is particularly strong competition for pineapple (from C6te dTvoire and Ghana), chilies (Cyprus), okra (Cyprus and the Far East) and mangoes (South America, Pakistan and India).

Table 13: Sources of supply to the UK horticulture market

Commodity Major supplying countries Normal availabUity

Avocado Israel October - May Canary Islands October - April Spain October - April Cypres October - May United States May-July Mexico May - September South Africa May - September Kenya August - October

Mangoes Mexico May - September Mangoes Venezuela April - July Brazil May - August Jamaica April - August India April - August Pakistan April - August Mali March - June cote d'lvoire April - July South Africa December - March Israel August • October Kenya October • July

Passion firuit Kenya All year round Brazil January - July cote d'lvoire June - December

Pineapples cote d'lvoire All year round Pineapples South Africa All year round Ghana All year round Kenya All year round

Strawberries Spain March - June France May - September Kenya September - April

Green beans Spain March - June Kenya All year round

Aubergines Cypres May - November Aubergines Netherlands March - November Canary Islands January - June Israel October - March Kenya All year round

Okra Kenya All year round Cyprus May - September Brazil October - May

Courgettes France April - December Courgettes Spain November - May Cypres Januaiy - November Kenya November - March Italy November - June Israel December - April

Cut flowers Netherlands (carnations, chrysantbemunis Channel Islands and roses) Spain and roses)

Columbia Israel Kenya

Asparagus fern USA Asparagus fern Kenya

Source: Chamber of Commerce 1989.

[c] Kenya's response

Kenya's response to this competitive challenge is to move up market into higher value produce. A start has been made on this with the move into green beans and, currently, strawberries. Other crops which may make their appearance (or increase from current modest levels) in the future are cherry tomatoes, asparagus, Galia melons and roses. Given both the competition and the constraints on air space, the bulkier products such as mangoes, pineapples, avocadoes and paw-paw will have either to be transported by ship or dropped from the range.

Ship transport raises its own problems because it requires large consignments which will ripen at the same speed en route. Kenya is not geared up at present to fulfil such requirements. The results of experiments to date have not been very encouraging [Chamber of Commerce, 1989:25].

Other areas in which Kenya needs to improve are on standards of quality and presentation. This, in turn, requires substantial investment in cold-storage and grading facihties at the level of both farm and point of export. It also requires improved packaging materials (which are perceived by the importers to be an integral part of the product). Supply schedules need to be improved and greater emphasis given to marketing [Chamber of Commerce, 1989:16]. Finally, there may be a move towards more 'all-year-round' exports of many items despite the very tight margins during the European summer.

A major challenge to which Kenya will have to respond in the near future arises from the changing nature of the horticulture trade in Europe. The large wholesale markets that have traditionally stood between producers and independent retailers are in decline. An increasing share of the horticulture trade is being captured by large supermarkets that deal directly with producers both within Europe and abroad. At present, the bulk of Kenyan horticultural exports are sold 'on commission' to the European wholesale markets. Dealing with supermarkets tends to remove some of the problems experienced by exporters, notably in respect of verifying prices and ensuring payment, as well as offering higher prices. But the supermarkets set stringent standards in relation to the quality of the product, the timing and consistency of supplies, and the packaging. As the supermarket share of the European horticultural market increases, so the scope for exporting in the current manner will decline.

III.4

Another significant and relatively 'new' export is canned pineapple. There is one firm in Kenya that is responsible for the bulk of canned pineapple exports. It is a subsidiary of Del Monte, the US-based multinational. Unfortunately, the firm declined to cooperate in any way with this study and so it has not been possible to analyse the sub-sector in detail. However, it is clear from Table 14 that canned pineapple exports increased steadily in volume terms during the 1970s, with a particularly large jump in 1975 when Lomi I came into effect, and have maintained that level during the 1980s.

In current value terms, canned pineapple exports by 1988 were about one-quarter of the level of fresh horticultural exports {Le. around 2.5% of total Kenyan exports).

Table 14: Exports or canned pineapple, 1971-1988

Volume Value ('000 tonnes) (K£ mUlion)

1971 10.8 1.0 1972 9.7 1.0 1973 13.4 1.5 1974 8.7 1.4 1975 20.4 3.6 1976 29.9 7.0 1977 45.3 10.5 1978 42.1 9.6 1979 41.1 9.3 1980 38.5 8.9 1981 40.9 12.0 1982 39.9 14.5 1983 47.8 20.9 1984 50.2 25.9 1985 44.5 24.4 1986 44.7 24.2 1987 43.0 25.8 1988 41.1 25.1

Source: Central Bureau of Sutistics. Statistical Abstract. Economic Survey, various issues.

Other processed fruit and vegetable exports are notable mainly by their absence. This is largely because of shortage of supply of raw materials (due partly to the higher prices received for fresh exports). The domestic market absorbs the whole output of the of the fruit and vegetable canning industry.

in.5 Clothing and Leather

Apart from horticulture, canned pineapple, and to a very small extent handicrafts, the only sub-sectors in which diversification has begun are clothing and leather. This appears to be the resuh partly of the characteristics of these industries and partly of the recent performance of the domestic market. Qothing is well recognised as an early manufactured export from developing countries and figures prominently in the diversification of other ACP states. Leather is a domestically and regionally produced natural resource. The recent shift towards exports may be due to the fact that the clothing and leather industries suffered a sharp drop in growth during the 1980s.

Between 1975 and 1980 the annual average real growth of value added in the clothing, textiles and leather sectors was 9.13%; in 1980-84 this rate slumped to 2.75% [Lewis and Sharpley, 1990:Table 7.5]. For manufacturing as a whole, by contrast, there was virtually no change in the rate of growth between the two time periods. Given the predominant role of domestic demand as a source of growth in manufacturing, this downturn in the clothing/leather sub-sectors may reflect a lack of buoyancy in the local market, and suggests that a search for exports could be a plausible response by producers seeking to revive their fortunes.

II1.5.1 Clothing

Although Kenya has a substantial clothing industry (with a gross product of K f 18.5 million in 1987), only one firm is able to export to the EC on a significant scale. It is alleged that this is because the domestic industry is, by and large, not competitive in international terms: it produces mainly for the domestic market, where it receives significant protection from imports.

The one exception is a firm that sells most of its output to the domestic market but as well has been increasing its exports slowly and cautiously. Established in 1968 in Kenya by its Indian principal, its initial objective was import substitution. The subsequent move into exports is now seen as a logical progression once the company had reached a certain size that made the domestic market too small. The company employs 2,800 workers, most of them men. This contrasts with the situation in Jamaica and with the stereotype of Third World clothing factories employing mainly women. The decision to employ men arises from the fact that the factory operates a 3-shift day.

The company produces both textiles and clothing, but only the latter is exported to the EC. The constraint on exporting cloth is less one of price than of quality. The company cannot guarantee the long lengths of completely unblemished cloth demanded by European clothing firms. Instead, with its lower labour costs, it is able to make clothes from slightly flawed material by cutting around the faults - a process that would be considered commercially unviable by a European company.

In 1988, the UK was its largest export market, followed by Zimbabwe, Uganda, India and Rwanda. Its exports are mainly trousers, which it sells at a rate of 9,000-10,000 pairs per month to the UK, plus small quantities of jackets and suits. The reason for exporting solely to the UK is a function of the nature of the product. The company began first to export synthetic trousers, for which demand is strongest in UK. Its first sales were in 1978/79, but exports really took off from 1986. It is now diversifying into a synthetic/wool mix, for which demand in mainland Europe is stronger. Hence, it envisages diversifying geographically and is beginning to look at the German market. Future plans are to diversify into knitwear.

The company aims for the middle of the market - not high fashion but not rock bottom prices either. It exports by either air or sea and has no major problems with transport. It sells always to a wholesaler, never direct to a retailer. The biggest constraint on expansion is believed to be the cost of imported equipment

which, of course, increases in terms of local currency as the Shilling is devalued. This would not matter if all output were exported, but the bulk of production remains destined for the domestic market. Increases in the local currency cost of inputs feed through into the local currency price of the goods produced.

No problems have been experienced with the EC rules of origin. No doubt this is because the company spins its cloth in Kenya and, in the case of synthetic/wool mixes, uses domestically produced wool The principal problem with penetrating the EC market is quality: the market demands high standards.

m.5.2 Leather

The bulk of Kenya's hides exports are still in relatively unprocessed form and, hence, do not count as non-traditional exports. However, government policy is to encourage a move into processed goods, and a tax is levied on unprocessed exports. Two companies have begun exporting processed leather on a significant scale. They have not yet reached the stage of exporting leather goods, such as gloves or handbags, but do export wetAilue and dyed leather.

Most of the exports are to the EC, largely because of traditional trading links. Being an animal product, there are veterinary requirements to be fulfilled. These vary between the member states (with Italy, Spain, Greece, Portugal and France being the most difiicult). Greece, for example, banned all imports of hides and skins for a year; Spain has a particularly cumbersome import regime. India and Pakistan are the most serious sources of competition.

nL6 Conclusions on Kenya

Until the mid-1980s, Kenya might have been taken as an exemplar by those critics of the Lom6 Convention and other trade preferences who argue that they fail to promote export diversification. As a coimtiy with a relatively well-developed production base, good manpower resources, favourable infrastructure and transport connections and a policy framework conducive to private sector growth, Kenya might have been expected at the time Lom6 I was signed to be in the forefront of ACP diversifiers. Yet, for the first decade of the Lomd Conventions, this failed to happen.

During the second half of the 1980s, however, Kenya has begun to diversify, in a fairly dramatic fashion and in a not totally expected direction. It has been the agricultural sector, rather than manufacturing, that has taken the lead. Within the space of a few years Kenya has become one of the EC's principal foreign suppliers for some horticultural products. In 1987, for example, it supplied one-third of all EC summer carnation and winter green bean imports from developmg countries, and three-quarters of EC summer green bean imports fi-om all sources.

Although it is never possible to make a definite link between a set of policies and an export flow, there is some reason to suppose that both supply- and demand-side factors have had an effect. Government policies, particularly in respect of the exchange rate, have helped Kenya to sell into the fiercely competitive horticulture market. At the same time, the Lomd concessions have been instrumental in allowing Kenya to take

advantage of this competitwe position. The Common Agricultural Policy for fresh fruit and vegetables is a major barrier to imports into the EC. The combination of, on the one hand, a generally protective regime that limits supply to the European market and, on the other, specific preferences for the ACP is a very favourable one in the following sense. The preferences give Kenya a competitive edge over other third country suppliers to the European market while the general protectionism means that prices in Europe are higher than they would be under a more hberal regime. As a result, Kenya is able to gamer some of the economic rent resulting from the restriction of supply. Of course, such positwe features of the regune could be more than ofEset if the CAP protection was a major constraint on the volume of Kenyan exports. However, the protectionist policies of EC states have been a constraint only on Kenyan exports of green beans and, in the future, might become a constraint for strawberries. For other products the principal constraints on increasing the volume of exports relate to difficulties of supply (including transport) together with stiff competition from other sources.

The challenge for the future is for Kenya to maintain and build upon the position it has established in a highly competitive market. It is clear that competition for Kenya's existing line of products is intensifying. To maintain its position in the horticulture sector, the industry will need constantiy to adapt its product range, production techniques and marketing systems. Not only will it have to move into new crops, but it will also have to come to terms with the increasing role of supermarkets in distributing fmit and vegetables. Otherwise, the export trade could collapse as fast as it has grown. At the same time, diversification needs to be extended into other sectors, notably clothing and leather.

A process of continuous adaptation to a remote market will present many practical difficulties for Kenya's, often smaU, export-oriented producers and traders. There is ample scope for a range of useful support measures to be financed through the aid provisions of the Lom6 Convention; the provision of supporting infrastmcture within Kenya, the supply of technical assistance in product development, help representing Kenyan export interests in the main markets, etc.

IV. ETfflOPU

The Ethiopian case study is necessarily more limited than the other four. This is for two reasons. First, Ethiopia exports a more Hmited range of non-traditional products than do the other countries given special attention. The major reason for including it in the sample of case studies is that it has been particularly successful for a least developed ACP state in diversifying. Second, the export of non-traditional products is undertaken almost exclusively by parastatal organisations. Hence, there are far fewer enterprises involved in the trade than is the case for the other countries studied.

The experience of Ethiopia may provide evidence relevant to other less developed members of the ACP group on the particular characteristics of diversification from a very weak economic base.

IV.l The Macroeconomic Framework

Ethiopia's economy has been disrupted severely during the period of the Lom6 Conventions by a combination of drought and political/military conflicts. These problems were particularly severe during the period of Lom6 I, but from 1979 imtil the mid-1980s there was a substantial recovery. This was brought to an abrupt halt by two severe droughts, in 1984/5 and 1987/8. In addition, in recent years, the military situation in a large part of the country has deteriorated sharply.

According to World Bank figures, Ethiopia is the poorest country in the world in terms of GNP per capita ($130 in 1987). Although it is primarily an agricultural economy (with agriculture contributing 42% of GDP in 1985/6), the manufacturing sector (8.5% of GDP) is quite significant [EIU, 1989]. Within manufacturing, food processing, textiles and beverages dominate; all are based on processing domestic materials. These three sub-sectors accounted for almost two-thirds of manufacturing output in 1986 [EIU, 1989]. Production is concentrated in and around the principal towns of Addis Ababa, Asmara and Dire Dawa.

The country has run a substantial balance of trade deficit throughout the 1980s. Since 1979 export earnings have stagnated even in nominal terms, largely because of the effect of the drought and military disruption on traditional staples of coffee and hides/skins. Ethiopia's most important export market is West Germany (with 29% of the total in 1985/6), followed by the USA (13%) and the Netheriands plus Japan (each with 9%).

IV,2 Relative Importance of Non-traditional Exports

It is quite apparent that Ethiopia has not diversified the commodity structure of its exports to any great degree. The principal non-traditional exports to the EC are horticuhural products (vegetables and flowers) and clothing. These still form a tiny proportion not only of total exports, but even of exports to the EC (Table 15). By 1986, the latest year for which figures are available, horticuUural products accounted for only 1% of exports to the EC, and clothing for a further 0.3%.

Table 15: Share of selected non-tradltkmal exports to EC in total exports, 1980-8<<

1980 1981 1982 1983 1984 1985 1986

Total exports (Birr million) 950.7 851.5 778.1 809.6 929.6 744.6 923.8

of which to EC n/a n/a 270.5 330.2 357.2 341.7 522.8

EC expon share (%) ot:'*^

Toul exports - - 35.0 41.0 38.0 46.0 57.0

Clothing - • 0.1 - - 0.1 0.3

Flowers/vegetables 0.2 - 0.2 0.1 0.2 1.0 1.0

Pulses 1.6 1.7 1.4 1.3 1.1 0.6 0.1

Notes: (a) Dau for total exports and exports to E C are not precisely comparable because different time periods are used and because of freight costs. The figures for total exports are derived from Ethiopian statistics and are for the Ethiopian calendar year (September-August), Le. 1980 e 1979/80. The figures are to.b. These data do not disaggregate sufficiently for the purposes of this ubie. The shares of clothing etc are derived lirom E C import data which are for the European calendar year and are ci.f.

(b) Share of each item in total exports.

Sources: National Bank of Ethiopia Quarterly Bulletin. Vol. 2 No. 2, 2nd Quarter 1986/87.

Eurostat.

Nonetheless, the growth of non-traditional exports has been quite rapid in recent years. The exports of horticulture and clothing have developed largely since 1984. Since the EC's share of total Ethiopian exports has increased substantially over this period, the growth of non-traditional exports is more rapid than may appear at first sight.

Non-traditional exports (defined as the 179 products analysed in this study) grew in current value terms by an annual average 4.5% between 1976 and 1987 (Figure 7). This overall figure conceals a decline during the troubled 1970s (-5.6% p.a. between 1976 and 1980) and a steady growth thereafter (10.7% p.a. between 1980 and 1987). There has also been a broadening of the range of non-traditional exports. In 1976, the only significant commodity was pulses. By 1987, clothing and horticulture had become prominent exports.

Figure 7: Growth of 'non-traditional' and total exports to the EC bi current value terms, 1976-1987 (Ecu million)

Non-trad.

V///////M Total

1976 1378 19B0 1982 1984 19B6 1377 1979 1381 1383 1985 19B7

Source: Eurostat

IV.2.1 Clothing

The Ethiopian textile and clothing industry is quite substantial. There are varying estimates of its relative importance: 20% of industrial production in 1985/86 according to the Central Bank (Table 16), or 14% of manufacturing value added according to the Central Statistical Authority (Table 17). Such differences simply underscore the likely margin of error in all such statistics; this margin of error should be borne in mind throughout this report.

In the period 1979/80-1985/86 the total value of textiles' production increased in constant prices by an annual average 3.7%. Within the clothing industry garment production increased particularly rapidly, at 16%, while knitwear grew more slowly (1.2%). As such, garment production grew faster than the average for all industries (Table 16). There are at least 60 enterprises invoh^ed in the textile and clothing industry (this is the mmiber of establishments covered by the industrial survey - Table 17).

The industry accounts for some 39% of total manufacturing employment. Interestingly, men and women are employed in roughly equal mmibers for the industry as a whole, but within clothing women predominate (Table 18). Wages appear to be slightly lower than the manufacturing average: whereas the industry accounts for 30% of male employment in manufacturing, it accounts for only 27% of the manufacturing wage bill; in the case of women it accounts for 61% of employment and 56% of wages.

Table 16: Value of industrial production In constant 1978/79 prices (Birr niinion)f''>

Av. annual 1979/80 1980/81 1981/82 1982/83 1983/84 1984IS5 1985/86 gmvlh(%)

Textiles (total) 303 324 332 341 344 351 376 3.7 Yarn (cotton acrylic) 64 65 66 68 68 73 79 3.6

Fabrics 165 175 183 184 177 170 187 Zl Garments 16 20 21 23 30 36 39 16.0

Knitwear 14 12 10 6 6 14 15 1.2 Leather 58 60 62 67 78 92 107 10.8

All industries 1213 1301 1368 1484 1472 1763 1863 7.4

Note: (a) Public and private.

Source: National Bank of Ethiopia, Ouanerlv Bulletin. Vol. 2 No. 2, 2nd Quarter 1966/87

Table 17: Value and costs of production In Ethiopian poUk and private clothing and textUe Industi7, 1985/86 (Birr IMW)

Number of eslablishments

Gross value of production

Industrial costs

Indirect taxes net

Censui Value Added at market at factor

prices cost

Spinning, weaving and finishing 20 335211 191216 34433 143995 109562 Knitting mills 23 13436 5287 1215 8149 6934 Cordage, rope and twine 5 36203 25712 2556 10491 7935 Ctothing (excl. footwear) 12 57209 37250 3981 19959 15978 Total textiles and dotbing 60 442059 259465 42185 182594 140409 Total all manufacturing (Birr million) 402 '2909.3 1583.2 428.1 1326.1 898.0 Textiles/ctothing as % of total 15 15 16 10 14 16

Source: Central Statistical Authoritv. Results of the Suivev of Manufacturine and Electncitv Industries 1978 E.C. (1985/86 G.C.): Statistical Bulletin 68 fAddls Ababa. January 19R9V

The export of clothing is undertaken almost exclusively by the National Textile Corporation (NTC), which is a government-owned parastatal. It has 19 factories (including three for jute) spread around the country and employs about 35,000 workers. As such, it is overwhelmingly the largest employer in the industry; in 1985/86 total employment in the industry was only 35,850 (Table 18). Only a very small part of its output is exported, roughly 5-10%. The main constraint on increasing exports is production capacity.

Given the level of domestic demand, exports could only be increased by the installation of new machinery. When account is taken of production by other, non-exporting companies, only 3% of clothing output is exported (Table 19).

Ethiopia is one of the few ACP countries able to satisfy the Lom6 rules of origin on woven clothing - a position it shares with Kenya but not with Jamaica. The country has both a supply of cotton and a weaving mill. It is able, therefore, to claim the Lom6 preferences on shirts and other woven garments.

Despite this advantage, the initial exports followed the normal ACP pattern and were of knitted goods. The NTC began exporting in 1981 with sales of T-shirts to West Germany. It had made contact with a German importer at a trade fair and the exports developed from this. There were sales for about two years and then the two sides failed to reach agreement on price and so the corporation began to export to East Germany instead. Subsequently a link was made with the Italian market when an Italian importer visited Ethiopia. By 1989 exports of garments were made to Italy and to East Germany, while finished fabrics were exported to the USSR.

The current shares of knitted and woven clothing in total Ethiopian exports of garments are not entirely clear. According to Table 19, the knitting output of the industry is consumed entirely in local sales. However, it is not clear whether the category 'Clothing' includes knitted cottonwear such as T-shirts and underclothes. It appears that it must do so since T-shirts remain the most important export item of the NTC. Recently the company has diversified into exporting shirts, swimwear, Bermuda shorts and ladies' aprons in addition to T-shirts. They also sell some grey fabric to UK, but have been unable to find a buyer for clothing there.

Exports to the EC have always been concentrated almost exclusively on underwear, ie. knitted goods. Since the rules of origin are not a constraint against exporting shirts and other woven goods, it must be assimied that the main limitation is to be found on the supply side , Le. total production is either inadequate or uncompetitive - or both. Although limited to one 4-digit product group, clothing exports have expanded rapidly. Sales of underwear to the EC increased in value by an annual average 46% between 1981 and 1987 (Table 20).

Table 18: Employment In Ethiopian textiles am 1 ctothing, IfSSm'"'

Number of eslablishmenis

Number of persons em EthioBJans Foreime

gloved T Total

Wanes (Birr -000) Ethiopians Foreigners

Male Female Male Female

Spinning, weaving and finishing 20 14258 12653 26911 39204 22608 KnitUng mills 23 483 928 4 1415 1019 1068 20 Cordage, rope and twine 5 •2923 1222 4145 4799 1532 -Qothing (exd. footwear) 12 1069 2310 3379 7/ion 3266 -Total textiles and clothing 60 18733 17113 4 35850 47622 28474 20 Total aU manti^uring 402 62771 28282 88 91141 173808 50611 2499 Textiles/clothing as % of total 15 30 61 5 39 27 56 1

Note: (a) Public and private enterprises.

Sourt«: Central Statistical Authoritv. Results of the Survev of Manufacturine and Electricitv Industries 1978 E.C n985/86 G.Cy. Statistical Bulletin 68 fAddis Ababa. lanuarv 1989V

Table 19: Domestic sales and exports of textiles and clothing, 1985/86'''> (Birr'000)

Revenue from sales lM:al Expon Total

Share of ejqiorts (%)

Spinning, weaving and finishing 321690 3262 324952 1

Knitting 13423 - 13423 0

Clothing 54399 1411 55810 3

Total 389512 4673 394185 1

Note: (a) Public and private.

Source: Central Statistical Authoritv. Results of the Survev of Manufecturine and Electricitv Industries 1978 E . C fl98S/86 G.CY Statistical Bulletin 68 fAddis Ababa. lanuaiv 1989).

Table 20: Exports or clothing to the EC, 1981-87 (Ecu '000)

NIMEXE Code Item 1981 1982 1983 1984 1985 1986 1987

6004 Underwear 123 352 146 151 375 1108 1166

6102 Women's outerwear 0 0 19 0 1 350 1

6103 Other men's 0 0 0 0 0 231 1125

6202 Linen, furnishings 0 3 49 24 0 0 0

Source: Eurostat

There is a fairly high import component in the value of exported goods. The government industrial survey for 1985/86 shows that for the knitting sector as a whole imports accounted for 20% of the cost of raw materials (Table 21). However, this almost certainly greatly understates the import cost of export items. Export goods have to be specially packaged and all the packaging materials are imported. The NTC estimates that roughly one-half of the total cost of ejqjort production is accounted for by imports.

Table 21: Cost of domestic and imported raw materials for clothing and textile Industry, 1985/86° (Birr '000)

Number of Local Imported Total Share of establishments imports (%)

Spinning, weaving and Hnishing 20 121514 43934 165448 27

Knitting mills 23 1451 3289 4740 69

Clothing 12 29108 7086 36194 20

Total 55 152073 54309 206382

Note: (a) Public and private.

Source: Central Statistical Authority, Results of the Survev of Manufacturine and Electricitv Industries 1978 E C fl985/86 G . C l : Statistical Bulletin 69 (Addis Ababa, Januaiy 1989).

IV.2.2. Fruit and Vegetables

The export of fruit and vegetables to the European market is undertaken solely by the Ethiopian Fruit and Vegetable Marketing Enterprise (EFVME). This was set up in 1980 and is a state trading organisation under the Ministry of State Farms. It is responsible to the Horticulture Development Corporation, which runs three other enterprises - all of them invohfed in the production of fruit and vegetables, together with processing for the domestic market. The EFVME is responsible for all aspects of marketing from the collection of the produce at the farm level. It undertakes quality control, domestic transport, packaging and delivery to the foreign market. Its labour force is relatively small; 230 permanent employees and 150-200 casual labourers. However, this is in addition to those employed on the state farms to produce the crops.

Horticulture exports are thtis a product of the state farms' system. State farms have been greatly favoured by the present administration. Not only do they occupy the prime agricultural land but, during the period 1980-85, they absorbed 40% of government expenditure on agriculture, received 76% of fertilisers available in the coimtry, 95% of improved seed and 80% of agricultural credit allocations. Yet they generated only 4-5% of agricuhural output [EIU, 1989]. As such, they have been widely perceived as representing a misallocation of resources, a view to which the Ethiopian government appears to have become more sympathetic in recent years. It is not possible, therefore, to assess the cost-effectiveness of agricultural production of horticuhural products for the export market. All that may be done is to note that, in a fiercely (X)mpetitive market, Ethiopia has been able to expand its sales, and that prices within the EC are relatively high because of the protection afforded by the CAP.

EFVME's export activities have grown significantly since it was established. At the start it exported 60-70 tonnes a year with a value of less than Birr 1 million. By 1988/89 exports had reached 3,500 tonnes, worth some Birr 9 million. When EFVME first began it had to transport all produce in ordinary trucks. The government has now invested in 28 refrigerated trucks and there are coohng facilities at the collection points. However, further infrastructure is required. Most urgent is a cold store at Addis Ababa airport; a 2,500 cubic metre store is currently imder construction. At present there are no facilities and the produce is stored in the direct simhght, leading to rapid deterioration if there is any delay in loading.

Current exports include the following flowers: statisce, delphiniums, dille, euphorbia, mulucella, carthamus and, just starting, astromaria. A range of other, higher value, flowers are on trial to see if they are suited to Ethiopian conditions. Vegetable exports include fresh green beans, sweet melons (Galia and Ogen), okra, a very small amount of asparagus, capsicums and, just starting, yellow onion. This variety is designed to achieve two objectives: to satisfy the importers' demand for a range of produce and to utilise fully aircraft space. Capsicums are a case in point. Spain is now a major supplier to the EC market so prices are very competitive. However, export remains worthwhile because they provide a filler to take full advantage of the capacity of the aircraft.

Although horticultural products still form a very small share of total Ethiopian exports to the EC, they are increasing fast (Table 22). Winter miscellaneous flower exports, for example, increased in nominal Ecu value between 1982 and 1987 by an annual average 62%, whilst exports of green beans increased by an average 84% p.a. between 1981 and 1987.

Table 22: Exports of flowers and firesh vegetables to the EC, 1981-87 (Ecu '000)

NIMEXE Code Item 1981 1982 1983 1984 1985 1986 1987

060319 Miscellaneous flowers 2 7 2 76 186 152 101

060361 Gladioli 0 100 0 86 409 112 0

060369 Miscellaneous flowers 1 237 223 621 2345 3364 2651

070145 Green beans 43 64 388 274 1048 1644 1674

070193 Sweet peppers 17 179 27 26 159 56 157

Source: Eurostat.

All horticultural products except yellow onion are exported by air. The choice of markets is determined largely by Ethiopia's air connections. All of the flowers are sold to the Dutch auctions, but vegetables are exported to Germany, France and UK, as well as to the Netherlands. Ethiopian Airlines collection points for charters are Amsterdam, Frankfurt and Rome, and so the tariffs for these destinations are the most competitive. There are 4-5 charters per day during the European winter, which is the only time the EFVME exports - prices are too low during the European suminer. Export is also by schedtiled flight to London and Paris (via Rome). However, tarife on scheduled flights are much higher than the rate on a fUUy-laden charter - Birr 1.5 per kilo. Freight costs account for aroimd half of the total cost of supply to the European market.

A major problem looming on the horizon for charters is that there have to be sufficient goods traveUing north to south to give the planes a full load in both directions. As foreign exchange constraints squeeze Ethiopian imports, so the capacity to export declmes.

As in Kenya, there are problems over packaging. At present all packaging materials are produced locally, but are not of the best quality for the European market. The exporters want the government either to renovate the sole corrugated paper factory in the country or allow them to import packaging materials.

Because the delivery mechanism is still fragile, Ethiopia has not achieved a sufficiently good reputation for reliability and quality to sell on quotation. Instead all exports are on consignment basis, ie. the importer sells at the best price he can get, deducts a commission and expenses, and remits the balance to the exporter. There are three disadvantages with this. One is that prices tend to be lower than on sales by quotation. Another is that the EFVME cannot always check whether the price quoted by the importer is a fair one. With flowers there is no major problem because all are sold at the Dutch flower auctions and they receive the computer printouts to check prices. For the fruit and vegetables, however, the BFVME has only the ITC news service as a guide. Finally, sale on consignment means that the exporter bears all the risk.

For the future, Ethiopia will have to contend, like Kenya, with the changing structure of the EC horticuhure market and the increasing role of the supermarkets. Competition to supply the declining wholesale market will intensify.

Another, more proximate, problem is that because of the altitude of Addis Ababa planes cannot take off fully laden, so some potential capacity is wasted. Until end-1989 the EFVME used Ethiopian Airlines Boeing 707s, with a capacity of about 31 tonnes, as well as scheduled services. But, from 1990, new noise restrictions in Europe require these to be replaced by leased aircraft with a hush kit attached. This will increase the freight cost. The restrictions should have been implemented from the end of 1988, but were extended for one year for those AC? states requesting a derogation; only Ethiopia took advantage of this opportunity.

IV3 Conclusion on Ethiopia

Given the broader economic problems that Ethiopia has faced since Lom6 I was signed, and the very brief period over which non-traditional exports have been made to the EC, it is not possible to draw hard and fast conclusions about the durability of the diversification process, its prospects, or the applicability of the Ethiopian experience to other least developed states. Nonetheless, it remains the case that Ethiopia has been able to begin exports of non-traditional products and, moreover, that these commodities are the very ones given the greatest effective trade preferences under the Lom6 Convention. It is also noteworthy that the markets for both clothing and horticulture are intensely competitive. Hence, the export success of Ethiopia, limited though it may be, is particularly encouraging as a sign that ACP states at all levels of development appear to have some capacity to break out of the stranglehold of traditional export commodities for which world demand is growing only slowly.

Moreover, despite its many unique features, Ethiopia shares some of the characteristics of the other two countries covered in this Working Paper. It is clear that both demand- and supply-side factors affect the level of exports. It is also clear that there is a range of supportive measures, that can be assisted through aid and technical assistance, that would relieve bottlenecks and enable the volume of exports to increase. It is feasible, therefore, to draw some general conclusions that apply to all three countries and, by imphcation, to a wider number of ACP states.

V. GENERAL CONCLUSIONS

These three case studies, together with those of Mauritius and Zimbabwe, were undertaken to illuminate the following questions:

• Have the countries diversified into genuinely non-traditional export commodities?

• How soundly based is this diversification?

• How far have EC policies, particularly within the Lom6 Convention, facilitated or hindered the process?

No small group of country studies can be taken as representative of a group as diverse as the ACT. However, the experience of the three countries studied in this Working Paper, together with Mauritius and Zimbabwe, may have a relevance that extends beyond their botmdaries. The statistical analysis undertaken in the first part of this research project indicated a wide range of ACP states involved in diversification, but a certain concentration on new products, notably horticulture and clothing. These five states have also diversified into this group of products.

Possibly one of the more important points to be drawn is that in all three countries covered in this Working Paper export diversification to the EC has occurred quite recently and, remarkably, that broadly similar conclusions have been fotmd also for Zimbabwe and, to a certain degree, for Mauritius (where an initial burst of diversification occurred during the 1970s but has been greatiy extended m the late 1980s). The reasons for this have been different in each case, and it is not suggested that there is any general force operating. Nonetheless, the uniform 'late start' has a relevance. It is that the actual diversification achieved by these states is greater than appeared from the statistical analysis, which covered the period only up to 1987 (after wMch the NIMEXE system of trade classification was superseded by the harmonised nomenclature, making detailed comparisons difficult).

This counsels the need to avoid genereilisations which are not continually reassessed in the light of each year's trade patterns. Much of the conventional gloom concerning ACP exports is derived from analysis which ends in the early 1980s (during a historically severe commodity price slump). It cannot be asstmied that the recent upturn will necessarily continue. All three of the countries considered in this paper are exporting into difficult markets; it does not follow that diversification will be as successful in the future as it has been in the recent past. But, by the same token, it is important to examine the experience of other ACP states in the most recent years to identify whether the diversification noted m the statistical analysis has been reinforced or undermined.

Another significant point is that all three states (plus Zimbabwe and Mauritius) have diversified into commodities which are particularly strongly protected in the EC and other OECD markets. In the cases of both clothing and horticulture, the protection is primarily in the form of non-tariff barriers that exclude absolutely some imports. Although the Lom6 Convention's trade regime is frequently described in terms of tariff

preferences, it would appear that its most potent features are in the form of preferential treatment concerning non-tariff barriers.

In none of the cases can an unambiguous causal link be drawn between the Lomd Convention and the development of non-traditional exports. A host of factors is at work to explain both the success in exporting these new products and the failure to export others for which Lome preferences are also substantial. Government policy in the exporting state is clearly a critical factor. In the cases of both Jamaica and Kenya, the lateness of the move into non-traditional exports can be explained in part by the unsupportive nature of goverrunent policies in the early period.

Nonetheless, there is some degree of circumstantial evidence to suggest that a link exists. The degree of preference that the ACP enjoy over third party exporters to the EC market is linked directly to the level of protection. The higher the level of protection for a particular commodity the greater the scope for the EC to provide preferences to the ACP. And we find that the export diversification of Jamaica, Kenya and Ethiopia has been precisely into those commodities on which EC protectionism is particularly strong and, hence, the Lomd preference is most marked. The result of EC protection is to Kmit supply onto the European market and,

therefore, to raise prices. The ACP, as privileged exporters, gain part of the economic rent in these high prices. Because the markets are competitive, the ACP have to be efficient if they are to sell at all; but, if they do reach this threshold efficiency level, their revenue is greater than it would be under free market conditions. In a sense, therefore, the combination of EC protection plus Lomd preferences provides something akin to 'infant industry' protection on the domestic market, but with the added element that competition is not suppressed completely. The danger that the economic rent will be absorbed in production inefficiency is reduced.

If there is some evidence that the Lomd preferences have been helpful, there is even more that imperfections in the Convention are a constraint on more rapid diversification. The rules of origin have been a clear impediment to some export diversification. As noted in the section on Jamaica, the rules of origin have been amended in Lomd IV. If the EC is willing to abide by the spirit of the new provisions it is possible that this barrier could be reduced. In addition to the rules of origin, the quotas that apply to many of the CAP concessions may become a constraint in the future. In the past the quotas appear to have been sufficiently large to accommodate most non-traditional ACP exports of CAP products (ie. excluding beef, rice, sugar and tobacco). But if, as seems possible, Kenya is joined by other ACP states as significant exporters to the EC, the group quotas could well become binding.

These conclusions have direct relevance to EC-ACP poHcy at a time when there is a general move towards liberalisation. To the extent that the GATT Round of Multilateral Trade Negotiations comes to a successful conclusion in December 1990 and leads on to the phasing out of the MFA, and to the extent that '1992' results in intra-EC liberaUsation, the ACP will lose some of their existing trade preferences. But the EC has it in its power to offer new concessions to replace the old. To the extent that the GATT Round fails to remove aU protection for temperate agriculture, the EC will retain scope for providing valuable extensions to the ACP trade regime during the course of Lomd IV in the form of new and enlarged concessions on CAP

products. In the case of some ACP countries, these new preferences may help to ofl&et the loss of tropical product preferences.

Moreover, there are opportunities within Lom6 to provide assistance to help overcome supply constraints. But, for them to be used, there has to be a recognition on both sides that this is a good use of aid.

The Lomd Conventions are broad, enabling documents, a catalogue of innumerable things that could be done rather than a precise action programme. The detailed focus is provided by each ACP countiy's National Indicative Programme (NIP), the negotiation of which provides the next step in moulding the Convention to the practical requirements of the 1990s. The idea that it can be used in support of trade appeara to have been understood by the Lomd partners: a joint declaration to the new Convention specifically mentions 'the need to give greater emphasis to the development of trade and services in the context of national and regional programmes of Community aid' [Part 2, Title X].

REFERENCES

Chamber of Commerce [1989]; The Kenya National Chamber of Commerce and Industry Report on First Specialized Mission on Horticultural Produce Organized bv the Kenya National Chamber of Commerce and Industrv Covering: Holland. Belgium. France and the United Kingdom. 12th to 28th September 1989 (Nairobi, mimeo).

Coopers and Lybrand [1988]; JNEC Export Qarment Industry Survey (Kingston, 2 May).

Economist Intelligence Unit (EIU) [1989]; EIU Country Profile 1989-90: Ethiopia (London, August 1989).

Harrigan, Jane [1989]; 'The conflict and crossing of conditionalities: a case-study of adjustment and stabilisation in Jamaica 1980-1988" (mimeo, 13 March).

JAMPRO [n.d.]; The Right Move for Your Apparel Concern (Jamaica Promotions Limited, Kingston).

JAMPRO [1989a]; Annual Report on Garments Sector 1988 (Jamaica Promotions Limited, Kingston, March).

JAMPRO [1989b]; Annual Report on Ornamental Horticuhure 1988 (Jamaica Promotions Limited, Kingston, March).

JNEC [1986]; Annual Report on Garments 1985 (Jamaica National Export Corporation, Kingston, March).

JNEC [1987]; The Third National Export Plan for Jamaica 1987-89 (Jamaica National Export Corporation, Kingston, January).

Kenya [1988]; Central Bureau of Statistics Statistical Abstract 1988 (Nairobi).

KiUick, Tony and Mwega, F.M. [1990]; Monetary Policy in Kenya 1967-88 (ODI Working Paper (forthcoming): Overseas Development Institute, London).

Lewis, Stephen Jnr and Sharpley, Jennifer [1990]; "Kenya", in Riddell, Roger et al Manufacturing Africa. (James Currey, London: Heinemann, New York).

Long, Frank [1986]; Employment Effect of Multinational Enterprises in Export Processing Zones in the Caribbean (ILO/UNTC Working Paper No. 42, Geneva).

McQueen, M. and Stevens, C. [1989]; 'Trade Preferences and Lomd IV: Non-traditional ACP Exports to the E C , Development Policv Review. Vol. 7 (Sage, London).

Steele, Peter [1988]; The Caribbean Qothing Industry: The US and Far Eastern Cormections. (The Economist Intelligence Unit, Special Report No. 1147, October, London).

UNILES [1987]; Techno-Economic Feasibility Study for the Manufacture of Wooden Furniture in Jamaica (UNILES for UNIDO, ljublyana, August).

Overseas Development Institute Regent's College Inner Cirele Regent's Park London NW1 4NS England Telephone: 071-487 7413 Fax: 071-487 7S90 Telex: 94082191 ODIUK